Schoff Etall U.S.-japan Final

WORKING PAPER
JULY 2021
A High-Tech Alliance:
Challenges and Opportunities
for U.S.-Japan Science and
Technology Collaboration
James L. Schoff, Douglas E. Rake, and Joshua Levy
A High-Tech Alliance:
Challenges and Opportunities
for U.S.-Japan Science and
Technology Collaboration
James L. Schoff, Douglas E. Rake, and Joshua Levy
© 2021 Carnegie Endowment for International Peace. All rights reserved.
Carnegie does not take institutional positions on public policy issues; the views represented herein are those
of the author(s) and do not necessarily reflect the views of Carnegie, its staff, or its trustees.
No part of this publication may be reproduced or transmitted in any form or by any means without
permission in writing from the Carnegie Endowment for International Peace. Please direct inquiries to:
Carnegie Endowment for International Peace

Publications Department
1779 Massachusetts Avenue NW
Washington, DC 20036
P: + 1 202 483 7600
F: + 1 202 483 1840
CarnegieEndowment.org
This publication can be downloaded at no cost at CarnegieEndowment.org.

Contents
Abbreviations vii
Summary 1
Introduction and Purpose 5
The U.S. Information Security System for

Science and Technology 8
The Japanese Information Security System for

Science and Technology 24
Alliance Cooperation in Theory and Practice 34
Options and Recommendations for Bridging the Gaps 38
Appendix A: U.S. Department of Energy Sensitive

Foreign Nations Control (Attachment G DE-AC08-36GO2830
Modification M1130), Reproduced 45
About the Authors 49
Acknowledgments 51
Notes 53
Carnegie Endowment for International Peace 67

Abbreviations
AI artificial intelligence
AIST (Japan) National Institute of Advanced Industrial Science and Technology
API Allied Prototyping Initiative
BARI Bilateral Academic Research Initiative
BIS (U.S.) Bureau of Industry and Security
CFR Code of Federal Regulations
COMSEC communications security
CoRe partnership (U.S.-Japan) Competitiveness and Resilience Partnership
CMI classified military information
CRADA cooperative research and development agreements
CSTI (Japan) Council on Science, Technology and Innovation
CUI controlled unclassified information
CWP Coalition Warfare Program
D10 a proposed club of democratic countries for secure technology supply chains
DCSA (U.S.) Defense Counterintelligence and Security Agency
DOE (U.S.) Department of Energy
DOD (U.S.) Department of Defense
DTAG (U.S.) Defense Trade Advisory Group
EAR Export Administration Regulations
FEFTA (Japan) Foreign Exchange and Foreign Trade Act
FOA funding opportunity announcement
vii
GSOMIA General Security of Military Information Agreement
ISA industrial security annex
ITAR International Trafficking of Arms Regulation
JAIC (U.S.) Joint Artificial Intelligence Center
JAXA Japan Aerospace Exploration Agency
JUSSTII Japan-U.S. Strategic Science, Technology, and Innovation Initiative
LAA Limited Access Authorization
LDP (Japan) Liberal Democratic Party
METI (Japan) Ministry of Economy, Trade and Industry
MEXT (Japan) Ministry of Education, Culture, Science and Technology
MIC (Japan) Ministry of Internal Affairs and Communication
MOFA (Japan) Ministry of Foreign Affairs
MOD (Japan) Ministry of Defense
NASA (U.S.) National Aeronautics and Space Administration
NBIB (U.S.) National Background Investigations Bureau
NDP (U.S.) National Disclosure Policy
NDPC (U.S.) National Disclosure Policy Committee
NISP National Industrial Security Program
NISPOM National Industrial Security Program Operating Manual
NISPPAC National Industrial Security Program Policy Advisory Committee
NIST (U.S.) National Institute of Standards and Technology
NNSA (U.S.) National Nuclear Security Administration
NOFORN no foreign nationals
NPUA nonproprietary user agreement
NSC (U.S. or Japan) National Security Council
OPM (U.S.) Office of Personnel Management
NSF (U.S.) National Science Foundation
QIS quantum information science
R&D research and development
RD restricted data
RIKEN (Japan) Institute of Physical and Chemical Research
S&T science and technology
SDS (Japan) Act on the Protection of Specially Designated Secrets
SPP strategic partnership project
TEPCO Tokyo Electric Power Company
VEP (U.S.) Vulnerabilities Equities Process
viii | A High-Tech Alliance: Challenges and Opportunities for U.S.-Japan Science and Technology Collaboration
Summary
In both Japan and the United States, there is a growing recognition that national security
and alliance security involve more than just military concerns and extend to new technology
areas and their economic effects. Looming over the two allies is a wide-ranging question of
how to enable greater science and technology (S&T) research collaboration for commercial
and defense purposes, involving such fields as artificial intelligence (AI), quantum informa-
tion science (QIS), cybersecurity, and space utilization. Competing domestic political and
economic considerations in each country make this a particularly complex challenge for
policymakers, who can be expected to treat the international sharing or sale of assorted tech-
nologies differently. Still, there is little doubt that it is in the mutual interest of Japan and the
United States to harmonize their approaches and find acceptable ways to enable deeper and
broader S&T collaboration—and to do so expeditiously.
This paper outlines a series of observations and recommendations intended to advance U.S.-
Japan S&T collaboration with national security implications.
• Although the United States and Japan have a productive S&T relationship, there are
significant gaps in terms of scale and scope when comparing their capacity to con-
trol and protect classified and sensitive information in dual-use technology areas.1
This is attributable to a combination of structural factors (such as the outsized role
of U.S. defense and intelligence budgets), legal differences, and cultural and histori-
cal factors such as greater sensitivity in Japan to blurred lines between military- and
commercial-application research.
1
• This situation imposes notable limits on the ease of bilateral defense trade, Japan’s
eligibility for partnership in certain sensitive U.S. government–funded research
activities, and the allies’ ability to jointly make use of private sector research and
development (R&D) with potential dual-use applications. As intensified technology
competition with China drives new public and private investment in all of these
areas, the allies will miss a strategic opportunity to leverage new resources if they do
not take steps to address these restrictions.
• A primary limiting factor for the participation of Japanese nationals in U.S.-

sponsored sensitive scientific research is the inability for Japanese investigators or
private sector engineers and scientists to receive domestic security clearances inde-
pendent of a national defense program. This could be remedied by an amendment
to Japan’s Specially Designated Secrets law similar to language in U.S. Executive
Order 13526 that expands the law to include “scientific, technological, or economic
matters relating to the national security.”
• Although security clearances are the first hurdle, a review of similar U.S. agreements
with other countries and existing military information agreements with Japan
reveals other areas that many U.S. officials still consider to be insufficient when
it comes to Japan’s security clearance process and information security regime. If
Japan wants to reach a substantially equivalent level with the United States and
conclude a new foundational information sharing agreement, it should consider
several steps including:
- clarifying its designated security authority in the government (possibly by

creating a “special secretariat” (tokubetsu no kikan) within the Cabinet Office
with interagency representation),
- creating a process for settling legal disputes in a classified manner, and
- developing a cadre of security professionals within the Japanese government

(similar to the GS-0080 class in the U.S. government).2
• The Japanese government should also consider extending a justification for “sci-
entific, technological, or economic matters relating to the national security” as an
amendment to the Foreign Exchange and Foreign Trade Act (FEFTA)—or some
similar action—that would strengthen the government’s ability to control private
exports of certain technology research information and private sector hiring of for-
eign nationals in select situations to address potential insider threat risks. Altogether,
this would raise the bar for industrial security in Japan and make it easier for the
allies to establish general licensing arrangements that could facilitate the exchange
of controlled goods and a wider range of sensitive information in a timelier manner
than is currently possible.
2 | A High-Tech Alliance: Challenges and Opportunities for U.S.-Japan Science and Technology Collaboration
• While there have been agreements that allow Japan and the United States to share
some sensitive and even classified information, there is a persistent view in the
United States that Japan needs a more robust system for properly handling and
protecting such information, particularly with respect to potential advanced S&T
collaboration.3 This paper explains why perceptions can often matter as much as the
tangible rules and procedures that exist when it comes to expanding information
and technology sharing.
• Japan requires a more holistic approach that considers high-priority security issues
in the development of a legal framework, with sufficient penalties, along with an
improved system that strengthens protocols and procedures for the classification of
information, technologies, systems, methodologies, and intellectual property. This
should include more secure ways of handling and protecting sensitive and classified
information, materials, and technologies, as well as other physical aspects such as
storage, limited access areas, and the like. Of equal importance is the selection,
vetting, and training of those individuals who may receive clearances, enabling
them to access classified information or materials. This points to the need for the
development of a more robust security culture in key institutions that will probably
include changes in some of the physical and digital facilities of these organizations.
• It is clearly in the interest of both Japan and the United States (as well as other close
partners) that Japan be seen as having the kind of information protection regime
that can enable an intensification of S&T collaborations that all parties agree is
essential to protect their common security interests. Thus, associated with some
Japanese reforms, there should be a public relations dynamic that calls attention to
these changes and reinforces this reputation publicly over time. In other words, the
Japanese government and private firms should strive to get the credit they deserve
for recent and future improvements.
• Of course, all of this must ultimately be planned and implemented by Japanese

parties to conform to Japanese law, its polity, and cultural norms. These reforms can
be accomplished in such a way to provide confidence in Japan’s system on the part
of the United States and other such allies. This will provide the essential foundation
for greater sharing of sensitive and even classified information, as well as facilitating
the growth of collaborative R&D in priority areas that have already been identified
as worthy of greater joint efforts.
• Additionally, in order to promote more synergy between public and private sector
entities and between the defense and commercial sectors, the allies should consider
establishing a bilateral public-private interdisciplinary body to work in support of
high-level U.S. and Japanese policymakers. This council could draw from top labo-
ratories and corporations to help policymakers leverage allied S&T collaboration in
new ways and delineate clear priorities amid fiscal constraints.
James L. Schoff, Douglas Rake, and Joshua Levy | 3

Introduction and Purpose
U.S.-Japan science exchanges and collaboration have a long history. Since the mid-1960s,
the two countries have worked together on natural resources and medical sciences, followed
by energy and outer space, and expanding more recently to include chemistry, computer
science, and telecommunications.4 The political and foreign policy context for this collab-
oration has evolved with time as perceived political priorities and opportunities ebb and
flow amid occasional direct competition. Throughout this half-century, however, there has
not been the kind of bilateral alignment that is on display today in terms of 1) a widely
perceived need for S&T collaboration with allies, 2) policymaker interest in pursuing this
collaboration, 3) specific areas of priority research focus, and 4) overall strategic goals for
such collaboration.
In early discussions, officials from the administrations of U.S. President Joe Biden and
Japanese Prime Minister Yoshihide Suga have highlighted three main goals for U.S.-Japanese
S&T cooperation: 1) mitigating climate change and its impacts, 2) being able to better
control pandemics, and 3) staying ahead of (or keeping pace with) China in a wide range
of critical technologies and domains of scientific discovery.5 The first two objectives lend
themselves relatively easily to multilateral collaboration and open information sharing. The
third goal is more pertinent in a bilateral context and suggests that more substantive allied
cooperation involving protected information is in the offing.
This concept of China as a so-called pacing threat has been referenced by U.S. secretaries of
defense serving both Biden and former president Donald Trump.6 It is also an underlying
premise for U.S. legislative initiatives that will change the way Washington funds and
5
directs high-tech research in the future.7 For example, one bill with broad support would
authorize a reconfigured National Science Foundation (NSF) to spend $81 billion over the
next five years, much of which is intended for research in AI, robotics, high-performance
computing, and other technologies.8
The case for U.S. S&T collaboration with Tokyo is clear. Japan is the third-largest investor
in R&D behind the United States and China, and Japanese companies fund more R&D in
the United States than any other country’s firms.9 Japan is also the third most prolific patent
filer in the world, behind China and the United States.10 Pooling resources, data, and talent
would benefit both allies. The question is not whether bilateral S&T collaboration is a shared
strategic opportunity but rather how to adapt it to keep up with new geopolitical realities
and the rate of technological change.
The allies are already moving in this direction. The Japan-U.S. Joint High-Level Committee
on Science and Technology Cooperation meeting in May 2019 called for greater bilateral
collaboration in areas such as AI, QIS, and outer space.11 Later in 2019, the governments
released a joint statement that outlined a more detailed agenda for bilateral collaboration on
QIS, building on previous bilateral initiatives in nuclear energy, high-energy density science,
and big data.12 Then, at their April 2021 summit, Biden and Suga announced the launch
of a new “partnership for competitiveness and innovation,” officially called the U.S.-Japan
Competitiveness and Resilience (CoRe) Partnership, which provides a framework for new
initiatives.13
Both governments recognize that scientific and technological advances in these and other
areas (such as cybersecurity, uses of the electromagnetic spectrum, and even critical infra-
structure protection and emergency response) can have a dramatic effect not only on eco-
nomic prosperity but also national security. As a result, U.S.-based research in such areas is
increasingly classified or export controlled, even though existing bilateral S&T cooperation
agreements—and, in many cases, certain aspects of domestic law—frequently do not allow
the allies to fully leverage their S&T talent and resources for mutual benefit. Policymakers
in both countries must consider ways to facilitate this cooperation if they want to realize the
expressed goals of the CoRe partnership.
Beyond bilateral initiatives, the United States and other nations are also looking to form
multilateral clubs (such as the UK’s proposed club of democratic countries [D10] or the
Global Partnership on AI) that involve standard setting or even R&D for certain cut-
ting-edge technologies that can help them outcompete China and enhance confidence in
the trustworthiness of their communications and innovation networks.14 Some of these
initiatives have explicit military connections (such as the U.S. Department of Defense’s
International Science and Technology Engagement Strategy or Allied Prototyping Initiative
[API]), while others seek to identify and more tightly control access to emerging, founda-
tional, or critical commercial technologies.15
Japan’s ruling Liberal Democratic Party (LDP) is similarly promoting the idea of a new
economic security law that could help Japan collaborate more with trusted international
partners, bolster innovation, and increase engagement in standard setting.16 The Japanese
government is already taking steps to enhance the country’s ability to protect valuable
information within the private sector and research community, which could facilitate its
participation in any multilateral clubs that might form around technology and innovation
imperatives.17 These trends suggest a need to harmonize allied approaches to research
conduct—or at least make them more compatible—so that the innovation ecosystem is able
to grow while maintaining mutual confidence in the protection of sensitive information.
Unfortunately, the complexity of domestic rules governing high-tech trade and research in
the United States and Japan (especially when it involves international partners) makes it
difficult to know what adjustments could be made that would meaningfully expand bilateral
S&T collaboration opportunities in a reliable way. This paper is intended to help policymak-
ers in both countries decide which reforms are most likely to effectively expand U.S.-Japan
collaboration on S&T research, development, and application in areas of common interest
and shared security priorities.
This paper’s objectives include:
• clarifying for both specialists and nonspecialists the policies and legal frameworks
that govern U.S.-Japan S&T research cooperation in sensitive technology areas,
including identifying both formal and informal barriers to bilateral collaboration;
• identifying areas of poor research protection compatibility and examples of missed

opportunities for bilateral R&D collaboration (with a specific focus on security
clearances and information security);
• recommending policy adjustments that can facilitate U.S.-Japan collaboration in the

near term and identifying issues that might require further study and debate over
longer time horizons; and
• building upon ongoing work by the Carnegie Endowment for International Peace,
Racke Strategies and Technologies, and others that is helping to design an improved
public-private process to take full advantage of the alliance’s collective S&T resourc-
es and apply them purposefully for mutual strategic benefit (known as the Japan-
U.S. Strategic Science, Technology, and Innovation Initiative, or JUSSTII).

The U.S. Information Security System for
Science and Technology
The U.S. Government’s Approach to Managing Classified Information
Because a considerable share of U.S. R&D on cutting-edge technologies is funded by the

federal government, often the results may be subject to government information controls.
Despite the military connotation, U.S. government information classification is not exclusive
to the Department of Defense (DOD). In fact, the three general levels of classification—top
secret, secret, and confidential—reflect the government’s evaluation that unauthorized
disclosure “reasonably could be expected to cause . . . exceptionally grave damage,” “serious
damage,” or “damage” to the United States’ national security.18 The existing classification
regime was promulgated by Executive Order 13526, signed by then president Barack Obama
in 2009.19 It delegates original classification authority to the heads of many government
departments and cabinet-level agencies,20 including individuals with job descriptions as
diverse as the head of the National Aeronautics and Space Administration (NASA), the sec-
retaries of defense or energy, the U.S. trade representative, and the Environmental Protection
Agency administrator.21
Executive Order 13526 also explicitly identifies “scientific, technological, or economic

matters relating to the national security” as a legitimate classification category. Indeed, many
kinds of research are considered pertinent to national security and are determined by their
respective departments and agencies to be classified as state secrets. This research is often
conducted at the national laboratories belonging to the Department of Energy (DOE) and
other U.S. government research institutions (though it also takes place at private research
facilities and with academic institutions). Further, such research is usually sponsored by the
DOD and/or the DOE. Though the order does state that “basic scientific research infor-
mation not clearly related to the national security shall not be classified,” there are few legal
guidelines that distinguish between basic research that does and does not relate to national
security. Regularly, material that is classified is not available to non-U.S. nationals unless a
specific determination is made to share that information.
Under certain circumstances, foreign nationals can be granted limited access to specific
types of classified materials. For example, under the DOD’s National Industrial Security
Program Operating Manual (NISPOM),
compelling reasons may exist to grant access to classified information to

a non-U.S. citizen. Such individuals may be granted a Limited Access
Authorization (LAA) in those rare circumstances where the non-U.S.
citizen possesses unique or unusual skill or expertise that is urgently needed
to support a specific U.S. Government contract involving access to specified
classified information and a cleared or clearable U.S. citizen is not readily
available.22
However, an LAA is restricted in several ways pertaining to scientific research exchanges.
An LAA does not grant the recipient access to information marked as top secret, restricted
data (RD), no foreign nationals (NOFORN), or communications security (COMSEC). It is
plausible that an LAA could be granted to a Japanese researcher or scholar for the purposes
of DOD- or DOE-sponsored research if they are granted an appropriate clearance in Japan.23
However, because of the ad hoc nature of the LAA program and its urgency requirement,
it is not likely to be appropriate for developing a sustained research collaboration program.
It also lacks a spirit of mutual benefit that would be necessary to sustain long-term bilateral
S&T cooperation related to shared security interests.
Over sixty-five years of defense cooperation, the United States and Japan have developed
a series of agreements and procedures for sharing sensitive and classified information with
each other, particularly when it relates to military information, defense equipment, and
information pertaining to nuclear energy.24 Currently, military and intelligence secrets are
shared between the Japanese and U.S. governments under the framework of the General
Security of Military Information Agreement (GSOMIA) signed in 2007.25 This agreement
allows for those who have received “personnel security clearance” from their respective
governments to participate in the transfer of classified military information (CMI). Each
side agrees to protect CMI in a “substantially equivalent” manner as the other, including in
storage, transmission, and vetting of those provided access (such as government or military
officials and contractors or sub-contractors) via a personnel security clearance system.
At present, the definition of CMI could conceivably include scientific information and
technology generated by Japanese or U.S. government entities, as long as that information is
clearly “defense-related.”26 More broadly, the U.S. government has developed a set of guide-
lines and procedures for sharing CMI and other sensitive information with other countries.27
Technical data, applied R&D, controlled unclassified information, and other components
of so-called international programs related to defense or national security are subject to the
National Disclosure Policy (NDP).28 Such disclosure is not limited to defense-related mate-
rial and could apply to a wider range of research linked to national security. However, there
are five basic disclosure criteria that must all be met before sensitive or classified information
can be shared with non-U.S. entities or personnel:
1. The disclosure must be consistent with U.S. foreign policy objectives vis-à-vis the
recipient country or region.
2. It must be consistent with U.S. military and security objectives.
3. The recipient nation will “afford the information substantially the same degree of
protection as the United States provides” (in terms of both intent and capacity).
4. The benefits to the United States of sharing the information should be equal to or
greater than the value of the information itself.

5. The disclosure must be limited to the minimum necessary to accomplish the sharing
objective.29
The third criterion is a major focus of this paper, so it is worth outlining some additional
details for how this determination is made. Regarding intent, the existence of a GSOMIA
with that country is generally considered sufficient for a favorable opinion. For the issue of
capacity, the U.S. National Disclosure Policy Committee (NDPC) sends officials to visit
partner nations, examines those countries’ security programs, and reports on their findings.
In addition, the U.S. intelligence community is also asked for its assessment of foreign
security programs and compliance with U.S. standards. Sometimes the NDPC also requests
special investigations of a foreign government’s practices regarding specific technologies or
capabilities. Endorsing sufficient capacity is sometimes incorporated into a foundational
information-sharing agreement, such as an industrial security annex (ISA) to a bilateral
GSOMIA.30
Overall, the NDPC places significant emphasis on how a partner nation decides whom
to trust within the government and how personnel security clearances are issued. Special
attention is paid to “access by contractors under third-party control or influence,” and
these contractors are required to have the same level of personnel security clearance as their
government-official counterparts (along with a “need to know”).31 The foreign country’s legal
framework must “bind contractors to the provisions” of their bilateral security agreement
related to CMI and other protected information, meaning that whatever criminal penalties
apply to unauthorized disclosure of the information by foreign government officials should
also apply to contractors.
Department of Defense Programs
In the interest of fostering international research collaboration on sensitive topics (and for
the purpose of advancing national security), the DOD runs and sponsors various research
programs involving other countries. For example, in 2018 the undersecretary of defense
for research and engineering, along with the UK Ministry of Defense, piloted the Bilateral
Academic Research Initiative (BARI) to advance basic research in the field of AI-augmented
human decisionmaking.32 Between the two governments, the team of researchers at academ-
ic institutions in the United States and in the United Kingdom received grants amounting to
just shy of $5 million. Though still a pilot program, initiatives like BARI serve as an import-
ant precedent for basic-research collaboration across national borders. Should it succeed, the
BARI pilot project could serve as a template for future collaboration between U.S. and allied
research teams.
The DOD sponsors and funds several other research collaboration efforts, which are gov-
erned by rules and procedures outlined by the NDP.33 Such cooperative R&D can take
the form of data and information exchange programs, engineer and scientist exchange
programs, specific cooperative projects (such as U.S.-Japan bilateral work on the SM-3
Block IIA missile), and foreign comparative testing to explore the compatibility of U.S.- and
foreign-made defense equipment.
The Coalition Warfare Program (CWP), for example, funds basic and applied research in
technologies to address “strategic technology gaps.”34 The CWP requires collaborative efforts
by U.S.-based and foreign-based teams and does not cover more than 50 percent of an inves-
tigation’s costs (thus requiring an ally to share the burden of additional research).35 Similarly,
the API is a program for pooling R&D funding and expertise with other countries.36 Both
CWP- and API-sponsored projects require government-to-government international agree-
ments. Consequently, on a case-by-case basis, it may be necessary for participating individu-
als and/or organizations to earn security clearances in their home countries. The U.S.-based
teams can, under certain circumstances, receive funding from both the CWP and API.
In fiscal year 2020, the DOD identified several priority areas for its investments, among
them quantum science, defensive and offensive hypersonics, directed energy, AI, and bio-
technology. Additionally, the fiscal year 2021 National Defense Authorization Act (NDAA)
allocated over $105 billion for research, prototyping, and associated evaluation programs.
Of that, over $8 billion was appropriated for basic and applied research, much of which will
be conducted through academic institutions and the armed services’ research laboratories.
Appropriations for offensive and defensive hypersonic technologies exceeded $1.5 billion;
directed energy research and prototyping programs received over $480 million; and quan-
tum research initiatives were allocated over $200 million.37
The NDAA also increased annual funding for the Joint Artificial Intelligence Center (JAIC)
that coordinates ongoing AI research—among many other activities—to $132 million.38 All
of these areas have also been identified by the Japanese and U.S. governments as high-pri-
ority areas for expanding bilateral science, technology, and systems R&D. Some of this
work, however, will fall outside of the CMI designation yet still involve access limitations
and dissemination controls, so cooperation may be difficult to accommodate under existing
U.S.-Japan information sharing agreements.
Department of Energy Programs
Though it is not often referred to or viewed as such, the DOE is effectively the United States’
department of science. Responsible for managing the network of national laboratories and
many of the researchers who work in those facilities, the DOE participates in and funds
both classified and unclassified research. In addition to managing research conducted at
national laboratories, the DOE also serves as a grant-awarding institution. It regularly issues
funding opportunity announcements (FOAs) to seek competitive submissions from the
general scientific community for funding in specific fields.
Some of the department’s classified work takes place in connection with the nuclear weapons
stockpile (such as the Stockpile Stewardship and Management Program or related initiatives),

or through contracts with the DOD, the intelligence community, and law enforcement
agencies in addition to other occasional government sponsors. Such work is consistently carried
out by the three biggest national laboratories: Lawrence Livermore National Laboratory, Los
Alamos National Laboratory, and Sandia National Laboratories. All three work under the
DOE’s National Nuclear Security Administration (NNSA). Similar work is performed by na-
tional laboratories administered by the DOE Office of Science, including Oak Ridge National
Laboratory, Pacific Northwest National Laboratory, and others.
It is important to note that these facilities work on sensitive and nonsensitive research proj-
ects alike—that is, both classified and unclassified work takes place at national laboratories
overseen by NNSA, the DOE Office of Science, and other DOE offices (like the Office of
Nuclear Energy). In many instances, this is possible because facilities within the campuses
are physically segregated such that classified and sensitive work can be appropriately protect-
ed and isolated in limited access areas away from researchers who may not have the appropri-
ate personnel security clearances.
Additionally, all work carried out at the national laboratories is governed by detailed clas-
sification guidance with no flexibility that is issued by the funding agency. Such guidance
may amount to few or no restrictions—or there could be an extensive set of instructions that
limit who inside and outside the laboratory is able to access project information. Moreover,
any time unclassified information is shared with people outside of the laboratory, every
PowerPoint slide, memo, or dataset must be clearly marked with an explanation for why that
information is allowed to leave the confines of that particular facility.39
With respect to international collaboration and unclassified work, the DOE is subject to
the same regulations as other agencies, including export controls. Indeed, chapter 25 of the
DOE Acquisition Guide explicitly notes that the fundamental research exemption identified
by National Security Decision Directive 189 (NSDD-189) “does not take precedence over
statutes.”40 That is, DOE-funded research materials, results, and documents ought to be
reviewed by the department’s Office of Classification prior to publication or dissemination
to ensure that they do not contain export-controlled or classified national security informa-
tion. The prepublication terms are generally negotiated on a case-by-case basis between the
department and the researchers with DOE funding.41
In some instances, as part of its collaboration efforts, the DOE may choose to establish a
strategic partnership project (SPP, formerly known as a work-for-others agreement) with
another federal or nonfederal entity.42 In this case, research work may be conducted in
tandem by DOE laboratories and other government agencies, private corporations, or foreign
entities. SPPs are required to meet the following criteria:
• The project must be consistent with or complement the identified laboratory’s

mission.
• The project must not adversely affect the performance of other laboratory programs.
• The project must not place the identified laboratory in direct competition with the
domestic private sector.
• The project must not adversely affect the future performance of the laboratory.
If the above criteria are satisfied, then the department may consider establishing an SPP.
Notably, laboratories cannot use government resources to conduct an SPP without an
appropriate and explicit allocation of funds to the project. Moreover, in the same spirit as
the above criteria, any laboratory that expends 20 percent or more of its regular operating
budget on SPP-related activities is subject to greater regulatory scrutiny.43 Intellectual proper-
ty and data resulting from an SPP can be protected as propriety information of the partner.
Additionally, the partner retains the right to elect the title to inventions and intellectual
property generated by an SPP, though the partner may opt for a limited government license.
Parallel to SPPs, the DOE also participates in cooperative research and development agree-
ments (CRADAs) with nonfederal entities. Unlike SPPs, CRADAs require the nonfederal
entity to provide funding for research or make in-kind contributions that may include, but
are not limited to, personnel, equipment, services, or intellectual property.44 The partner
entity may pay up to 100 percent of associated costs. SPPs, by contrast, may simply involve
a nonlaboratory entity paying the laboratory to conduct research work. CRADAs allow the
partner entity and the laboratory to share generated intellectual property, though each party
retains title to its own inventions. Oftentimes, the partner has the option to negotiate an
exclusive license to the laboratory in question. Data generated as a result of the cooperative
investigation can be protected from public disclosure for up to five years.45
Finally, third-party researchers and institutions can sign a nonproprietary user agreement
(NPUA) with many of the national laboratories. NPUAs allow third-party “users” to
utilize DOE or laboratory equipment for experiments, provided the users pay for their own
experimental costs.46 Under this structure, users generally commit to openly publish the
results of any investigation. Consequently, NPUAs are not used for classified work and users
are frequently academic institutions. Under an NPUA, the DOE retains unlimited rights
to access data generated by the investigation, but the user retains title to the underlying
intellectual property. If the users wish to keep the results and resultant intellectual property
of the investigation as proprietary information, they can negotiate access with laboratories
or the department on a full cost recovery basis. User agreement proposals are evaluated by a
merit review process.47
The DOE has collaborated with Japanese entities on advanced research topics in the past
and continues to do so. In June 2021, the DOE and Japan’s Ministry of Education, Culture,
Sports, Science and Technology (MEXT) signed a project arrangement to promote R&D
on quantum communication, computing, and other aspects of QIS.48 The FOA was facilitat-
ed by a series of overlapping and concentric international cooperation agreements. From the
most fundamental to the most accessory, the legal frameworks included:

• the 1988 U.S.-Japan Agreement on Cooperation in Research and Development in
Science and Technology;
• establishment of the bilateral Civil Nuclear Energy Research and Development

Working Group in 2012;
• the 2013 DOE-MEXT Implementing Arrangement Concerning Cooperation in

Research and Development in Energy and Related Fields;
• a 2015 DOE-MEXT Project Arrangement concerning high energy physics;49
• a 2015 Implementing Arrangement Between the Department of Energy and the

Ministry of Economy, Trade and Industry of Japan Concerning Cooperation in
Research and Development in Energy and Related Fields; and
• a 2019 DOE-MEXT Project Arrangement concerning high energy density science.
During research collaboration activities and visits to department facilities, foreign individu-
als and entities are supposed to be screened by the department according to the Science and
Technology Risk Matrix.50 Managed by the DOE Federal Oversight Advisory Body, the
S&T Risk Matrix evaluates the sensitivity of the subject matter and the nationality of the
foreign individual/entity to determine if or what kind of access should be allowed to facil-
ities and information.51 Thus, some “sensitive country foreign nationals” may be permitted
to access certain facilities or participate in certain programs but not others depending on
the level of sensitivity. For instance, a Chinese researcher might be barred from visiting an
NNSA-managed laboratory or participating in stockpile stewardship but might nevertheless
be allowed to participate in research pertaining to solar photovoltaic cells.52
It could be worthwhile to examine U.S. arrangements with Israel in more detail, given the
two countries’ close collaboration on science. Despite Israel’s inclusion on the department’s
sensitive country list, the relationship occasionally involves classified information.53 The same
is true to some extent with India, which is also on the sensitive country list but recently
concluded an ISA with the United States that allows for a wider range of defense technology
cooperation. The key difference in these cases—in terms of why Washington allows for more
permissive information sharing—is that both countries are recognized for having a well-de-
veloped and robust system for protecting classified information, including an effective
process for vetting and training people or organizations that are granted clearances.54
Government-Supported Unclassified Science and Technology Research
The NSF is an independent agency of the U.S. federal government that funds basic and
applied physical and social scientific research. As an independent agency, it does not report
to a higher-level government entity as, for instance, the National Institutes of Health reports
to the Department of Health and Human Services.
Although the NSF does collaborate with other government institutions, the vast majority of
its funding is directed toward colleges, universities, and other academic entities.55 In fiscal
year 2020, 80 percent of the NSF’s $7.9 billion budget was awarded to educational institu-
tions for research to be conducted by their scholars. A further 13 percent was awarded to
small business, nonprofit institutions, and private industry.
While the NSF funds work in many areas of basic and applied sciences that may be deemed
sensitive, the agency’s policy states that “NSF grants are intended for unclassified, publicly
releasable research.”56 As a consequence, NSF grantees are not granted access to classified
information by proxy of winning a funding award.57 However, the NSF notes that, in line
with other statutes concerning the protection of classified information, if a grantee’s research
produces information or results that may be deemed classifiable, it is incumbent upon the
grantee to notify the appropriate NSF program officer or federal government representative
promptly.
As competition between the United States and China intensifies, the U.S. Congress and
executive branch agencies have become increasingly concerned about protecting sensitive
U.S. research efforts from Chinese exploitation. In response, they have placed new restric-
tions on who can be eligible for NSF funding and pressed for tougher scrutiny to maintain
so-called research integrity. This term goes beyond the underlying veracity and soundness of
the research itself to include ensuring that researchers have no undisclosed or compromising
connections to rival nations, such as China.
A 2019 “Dear Colleague” letter from the director of the White House’s Office of Science
and Technology Policy, for instance, stressed the importance of establishing and coordinat-
ing disclosure requirements for researchers applying for federal funding.58 New guidelines
and rules for stricter enforcement of conflicts of interest, participation in foreign talent
recruitment programs, and affiliations with other research institutions have been prescribed
for the NSF in a December 2019 report by JASON, an independent scientific advisory
group.59 Thus, although the NSF does not conduct personnel security clearance background
checks in the traditional sense, the enhanced disclosure requirements serve as a soft security
measure.
Additionally, since the publication of the JASON report, the NSF has created the Research
Security Strategy and Policy Group, led by a new chief of research security, to work with the
Office of the Inspector General on auditing these disclosures for potential grant fraud.60 An
initial audit of past grants found up to 8 percent contained anomalies (such as subsequent
publication of NSF-funded research crediting another entity in addition to the NSF that was
not disclosed on the NSF grant proposal).61 This produced several cases that have subse-
quently been referred to the Federal Bureau of Investigation (FBI) for further investigation.62
In the same vein, the NSF ended foreign participation in its Intergovernmental Personnel
Act (IPA) assignment program, whereby the NSF would assign outside personnel to rotate
through various research institutions, including government agencies and private and public

universities.63 Now, all such rotating personnel must be U.S. citizens. NSF employees are
also barred from participating in foreign talent recruitment programs, though this policy
presently does not extend to include NSF grantees who are employees of third-party
institutions.
Despite its increasing focus on research security, the NSF continues to maintain robust
international cooperation programs, although it is possible that information protection
requirements will be enhanced in the future.64 Among several hundred international coop-
eration awards funded by the NSF, many are in conjunction with Japanese researchers and
agencies. Projects include researcher exchanges on topics as diverse as exascale computing
technologies for data analytics and deep learning; biobased materials science; investigations
of computational analysis of solid oxide electrolysis for energy storage; and network-of-net-
works organization for creating roadmaps for future investigations and demonstrations of
quantum technologies.65
These projects collectively receive millions of dollars in NSF funding, which Japanese
agencies usually match by funding their own researchers who are collaborating with U.S.
partners. Information is supposed to flow freely between these bilateral teams, but the NSF
is considering the inclusion of stricter information protection requirements on grantees for
experiment data, prepublication information, and the grant proposals themselves.66 Congress
is pushing for this as well.67 Any mismatch in information handling protocols could compli-
cate research collaboration. It would be especially unfortunate if this problem arose just as
Congress, which is looking to vastly increase funding for the NSF, has singled out Japan as a
priority innovation partner.68
Private-Sector Considerations and Classified Patents
For all the focus on how government funds are spent on basic research in the United States
(and efforts in recent years to increase that spending), the private sector supports most S&T
research (see figure 1). At one point in the 1960s, during the Cold War, federal government
R&D spending peaked at nearly 2 percent of the country’s gross domestic product. In recent
decades, though, the ratio has dropped below 1 percent while private industry spending has
grown. Nondefense high-tech firms are the most prolific spenders on R&D (see figure 2),
although only a relatively small portion is dedicated for basic research and shared publicly.
The huge amount of private sector money spent on R&D means that governments must
continue to stay connected to the private R&D infrastructure and help companies protect
their valuable intellectual property spanning a wide range of end uses.
FIGU RE 1
U.S.1.R&D
Figure U.S.Funding Source by
R&D Funding by Sector
Sector
2.5
FEDERAL
2.0
SH A R E O F GDP (% )
1.5
BU SINESS
1.0
0.5
OTHER
0.0
1953
1955
1957
1959
1961
1963
1965
1967
1969
197 1
1973
1975
1977
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
2011
2013
2015
2017
2019
SOURCE: "National Patterns of R&D Resources: 2018–19 Data Update," April 9, 2021, https://ncses.nsf.gov/pubs/
Source: National Science Foundation
nsf21325#data-tables
Note:NOTE:
TotalsTotals are approximate
are approximate duetotorounding.
due rounding. "Other"
“Other”includes funding
includes for U.S.
funding R&D R&D
for U.S. by nonfederal government,
by nonfederal government,
institutions
institutions of higher
of higher education,and
education, and other
othernonprofit
nonprofitorganizations.
organizations.
Private sector technology R&D is ostensibly beyond the reach of U.S. government classi-
fication control. There is, however, a process whereby the government can issue classified
patents, which serve similarly to restrict access and increase penalties for unauthorized
disclosure. In 1951, Congress passed the Invention Secrecy Act.69 If the commissioner for
patents of the U.S. Patent and Trademark Office determines a patent application contains
intellectual property that would threaten U.S. national security should it become public,
the commissioner can issue a secrecy order to the patent filer. This restricts the ability of
the patent filer to publish the intellectual property enclosed in the patent application. That
is, the secrecy order applies to the subject matter itself and not merely the contents of the
application. Usually, if the patent is subject to a secrecy order, the U.S. Patent Office will not
issue the foreign patent license that is regularly issued upon the receipt of a patent applica-
tion. This limits the patent filer’s ability to seek patent protection in foreign jurisdictions.
Generally, the commissioner for patents determines whether a patent contains such infor-
mation in consultation with officials in branches of the armed forces, the National Security
Agency, the DOE, NASA, and the Department of Justice. Potentially secret patent applica-
tions are checked against the Patent Security Category Review List (PSCRL), which outlines
the types of technologies that may be deemed national secrets. The PSCRL however, is a

Figure 2. U.S. Tech Company R&D Spending Compared to Defense Contractors
FI G U R E 2
U.S. Tech Company R&D Spending Compared to Defense Contractors
FI GU
U.S. R E 2FIRMS
TECH
U.S. Tech Company R&D Spending Compared to Defense Contractors
A M A ZON $22.6
U.S. TECH FIRMS
AA
LPHA BET $16. 2 $22.6
M A ZON
INTEL
A LPHA BET $13.1 $16. 2
M IC ROSOFT $12. 3
$13.1
INTEL
A PPLE $11.6
$12. 3
M ICROSOFT
0 5 10 15 20 25
A PPLE $11.6
2018 R& D B U DGE T (B I L L I O NS O F U . S. DO L L ARS)
U.S. DEFENSE
0 FIRMS 5 10 15 20 25
BOEING $3. 2
U.S. DEFENSE FIRMS
UNITED $2.4
$3. 2
BOEING
TECHNOLOGIES
$1. 2 $2.4
LOCKHEED
UNITED
M A RTIN
TECHNOLOGIES
RAYTHEON $0.7
$1. 2
LOC KHEED
M A RTIN
NORTHROP $0.6
$0.7
RAYTHEON
GRUM M A N
0 5 10 15 20 25
NORTHROP $0.6
GRUM M A N 2018 R& D B U DGE T (B I L L I O NS O F U . S. DO L L ARS)
SOURCE: James 0Manyika, William H. 5McRaven, and Adam10Segal, “Innovation and15 National Security: 20
Keeping Our Edge,” 25
Council on Foreign Relations, updated September 2019, 29, https://www.cfr.org/report/keeping-our-edge.
SOURCE: James Manyika, William H. McRaven, and Adam Segal, “Innovation and National Security: Keeping Our Edge,”
Council on Foreign Relations, updated September 2019, 29, https://www.cfr.org/report/keeping-our-edge.
Source: James Manyika, William H. McRaven, and Adam Segal, “Innovation and National Security: Keeping Our
Edge,” Council on Foreign Relations, updated September 2019, 29, https://www.cfr.org/report/keeping-our-edge.
classified document.70 A 2009 Freedom of Information Act request revealed that, at that
time, twenty-two “groups” of technologies were considered subject to potential classification.
However, the remainder of the document was redacted, giving little indication to those
applying for patents if their inventions could be subject to secrecy orders.
There are three types of secrecy orders that the commissioner can issue:71
• Type I: Foreign patent filings may be permitted for certain countries. Such an order
permits patent applications that pertain to some military and space technologies.
Notably, because the United States and Japan have reciprocal security arrangements,
Type I secrecy orders may allow for U.S. inventors to apply for patents in Japan.
This type of order is the most permissive and is intended to control the dissemina-
tion of the subject information while still allowing for its utilization.
• Type II: The contents of the patent filing should be treated as if it were classified or
classifiable technical information. As a result, Type II orders allow for the disclosure
of technical information in accordance with NISPOM. Type II orders are usually
issued to those who have or are subject to a DOD security agreement.
• Type III: The contents of the patent filing should be treated as if they were classified
or classifiable technical information. Type III orders generally prohibit the disclo-
sure of the subject matter without the express permission of the commissioner. This
type of order is issued when one of the above two types would not be appropriate.
Type III orders are usually issued to those who do not have a DOD security
agreement.
Secrecy orders last for a period of one year and can be repeatedly renewed by the commis-
sioner. At the end of fiscal year 2020, 5,915 patents were subject to secrecy orders, forty-five
of which were newly issued.72
There are also other private sector considerations related to information security and in-
ternational collaborations. Increasingly, the U.S. government is collaborating with private
industry to improve cybersecurity practices and protect Americans’ personal information
and trade secrets. However, because the U.S. government seeks out vulnerabilities to conduct
its own intelligence gathering and cyber espionage activities, the various government agen-
cies that discover, develop, and use so-called zero-day exploits are required to balance the
imperatives of its classified missions and interests against the public interest in disclosing the
existence of such exploits to developers who could patch these vulnerabilities.
To balance these competing interests, the government established the Vulnerabilities

Equities Process (VEP) in 2008 (last publicly updated in 2017).73 If the constituent mem-
bers of the Equities Review Board determine that the disclosure of a discovered, classified
vulnerability will satisfy a certain public interest without damaging military, operational,
intelligence, commercial, law enforcement, or diplomatic interests, the vulnerability can be

disclosed and reported to a software developer to be rectified. The Equities Review Board
can also choose to inform “allied government entities of the vulnerability at a classified
level.”74 In the case of Japan, however, it is not clear how Japanese authorities could share
relevant information with certain private companies that might need to take countermea-
sures. As this paper explains later, Japan’s information security system does not extend into
the private sector as extensively as does the U.S. system.
Additionally, because of the United States’ robust defense and civilian research infrastruc-
ture and bases, many firms from around the world choose to locate R&D facilities in the
United States. Though the U.S. government does award research contracts involving classi-
fied information to non-U.S. firms, companies that are subject to foreign ownership, control,
or influence (FOCI) are required to satisfy certain conditions to ensure the security of that
information.75 A firm is considered to be under FOCI if it satisfies any of a number of condi-
tions, including but not limited to: large tranches of equity or debt obligations being held by
a foreign entity or entities; a non-U.S. citizen or citizens being on the board of directors or
in senior executive or management positions; a foreign entity or entities having the power to
appoint or control the board of directors; significant revenue or income reliance on foreign
markets or entities; and other personnel or financial involvement with foreign entities.
Firms that are deemed to be subject to FOCI are required to pursue at least one of a number
of remedies to mitigate that FOCI.76 A FOCI firm may pursue a board resolution, the least
onerous of the possible remedies, to ensure that foreign entities that may control or exert
influence on the firm do not have access to any classified information. Because this is the
least restrictive mitigation instrument, it is unavailable to FOCI firms that have foreign
nationals on their board of directors. Firms can opt to create a special security agreement
(SSA), whereby a largely independent, wholly U.S.-subsidiary firm is established to create
distance between the foreign entities of potential concern from the new firm that would
handle the classified contract and information. It must be demonstrated to the Defense
Counterintelligence and Security Agency (DCSA), which is responsible for FOCI firm
administration, that the new entity would be financially viable.77 Finally, a FOCI firm could
opt to create a voting trust or proxy agreement, whereby the foreign entities place several
of their authorities over the firm (such as voting rights, corporate management, mergers, or
declarations of bankruptcy) in a trust of DCSA-cleared U.S. citizens. Because this, in effect,
requires the foreign entity to relinquish control of the firm, it is the most restrictive option
and the one that is pursued least frequently.78
Personnel Clearance Processes in the United States
The process by which the U.S. government evaluates its military and civilian personnel—as
well as private citizens—to issue security clearances has evolved over time, but it has suffered
repeated criticism for being inefficient. For example, unsatisfied with the performance of the
government’s background investigation process, Congress and the Obama administration
pushed reforms in 2016 that consolidated those activities in the National Background
Investigations Bureau (NBIB) within the Office of Personnel Management (OPM). At that
time, the backlog for investigations was over 570,000 cases, up from 328,000 cases the
year before.79 But NBIB was unable to turn the situation around, and the backlog grew to
700,000 in 2017 with an average investigation period of 450 days.80 The security clearance
backlog peaked at 725,000 in 2018. By then, there were calls in Congress for additional
reforms.81 A massive cybersecurity breach at OPM in 2018 sealed the fate of NBIB, and its
operations were transferred to DOD starting in 2019.82
Since 2019, the newly established DCSA (created by Executive Order 13869) combined NBIB
with the Pentagon’s own Defense Security Service to assume responsibility for roughly 95
percent of all federal government background investigations.83 That amounts to about 2 mil-
lion background investigations each year.84 Under DCSA, the personnel security investigation
mission was broadened to include insider threat analysis and a continuous evaluation and vet-
ting program, with an estimated 675,000 industry employees enrolled in continuous vetting.85
DCSA also has an industrial security mission in its organization’s roots, and it oversees im-
plementation of the National Industrial Security Program (NISP) and NISPOM (established
via Executive Order 12829 in 1993 and subsequently updated by Executive Orders 12885 and
13691).86 The DCSA director reports to the undersecretary of defense for intelligence.
The NISP was created to make sure that U.S. industry is properly prepared and effectively
carries out its responsibility to protect classified information in its possession when perform-
ing work on government contracts. DCSA clears over 12,000 contractor facilities for hosting
classified materials belonging to the DOD and thirty-three other federal agencies, including
NASA, the State and Commerce Departments, the NSF, and others.87 In the first half of
fiscal year 2020, DCSA processed 130,000 industry requests for personnel investigations. It
oversees 844,000 contractors with clearance eligibility, of whom about 455,000 are enrolled
in a continuous evaluation program.88
The NISP is supported by a public-private advisory committee called the National Industrial
Security Program Policy Advisory Committee (NISPPAC), which combines government
agency representatives with those from private companies (like General Dynamics or
ManTech) and research entities (like the Massachusetts Institute of Technology or the
RAND Corporation) who review and recommend changes to industrial security policy.89
NISPPAC is a forum where relevant government agencies can update industry about pro-
posed modifications to current practice and where industry can seek clarification, push back
against proposals, or suggest different priorities based on their understanding of the needs in
the field and compliance challenges.
Controlled Unclassified Information
In many settings, the U.S. government has an interest in restricting access to certain types
of information even where it cannot be classified. Since 2010, when Obama issued Executive
Order 13556, such information can be deemed as controlled unclassified information

(CUI).90 The original intention of the order was to harmonize various unclassified dissem-
ination control standards that were being used by the many agencies of the U.S. executive
branch. Agencies throughout the government are each individually responsible for determin-
ing what is considered CUI, but standards for the handling of that information are meant to
be uniform.
The bulk of CUI in the United States has no connection to national or economic security. It
frequently covers personal information, personnel records, health records, or other informa-
tion protected by law from public release. Important exceptions to this rule include critical
infrastructure, nuclear, intelligence, and export-controlled research.
As part of the United States’ export control regime on sensitive technologies, the Export
Administration Regulations (EAR) and International Trafficking in Arms Regulations
(ITAR) are the two primary statutes that govern R&D. The latter pertains to military and
aerospace technologies more narrowly and is administered by the State Department, while
the former concerns almost all other sensitive technologies (such as communications, non-
military nuclear, industrial, biological, or chemical) and is managed by the Department of
Commerce. The DOD and other agencies also play an active role in the U.S. export control
regime.91
So-called export-controlled research largely falls under the purview of the Department of
Commerce’s Bureau of Industry and Security (BIS), which can determine such research
to be CUI and subject to export control licensing requirements. This has implications on
research that may be conducted by foreign nationals and involve foreign funding.
Most basic research that is conducted at universities falls under the fundamental research
exemption and is subject to less restrictive (or no) controls on the export of sensitive technol-
ogies or software.92 If none of the following conditions are violated, then scientific research
can be regularly conducted by foreign nationals and with foreign funding unless explicitly
prohibited by Title 15 of the Code of Federal Regulations (CFR), Chapter VII §734.6-10.
The conditions that must hold include the following:
1. The principal investigator(s) must intend to publish the results of fundamental

research such that they are regularly available to the wider scientific community.
Such publications include books, academic journals, and the like. That is, the
results of such research must not be considered proprietary to a firm or other such
organization.
2. There must not be any ex ante limitation by research sponsors of any publication
of research results or quid pro quo side deals between the sponsor and principal
investigator(s).
3. Federally funded research must pass a prepublication process to determine whether
any sensitive research is contained in the research results. Such research can be
found not to be eligible for the fundamental research exemption at the discretion of
the reviewing agency.
4. There must not be any material transfer of goods that are the products of fundamen-
tal research across borders. Only the results may be published and made available to
the broader scientific community.
5. The fundamental research must take place within the borders of the United States.93
For research that does not fall within the above parameters—particularly corporate R&D—
resultant technology and software may be deemed CUI and be subject to certain export
controls by the BIS.94 In the case that research and its outcomes are determined to be CUI,
international information sharing is still possible. CUI is eligible for the marking REL TO
[country], which means release to nationals specified, and interested parties can apply to the
BIS for export licenses for such information. Such applications are considered on a case-by-
case basis, and there still can be a need-to-know requirement in some situations.95 Moreover,
the disclosure of CUI to nonexecutive branch entities should be regulated and managed by a
written agreement, and disclosure of CUI to foreign entities should also be limited according
to a written agreement and other specific criteria.96
In order to ease the process of arriving at an agreement, international research partners

can adhere to the same standards for protecting CUI as specified by the Department of
Commerce’s National Institute of Standards and Technology (NIST). NIST’s Protecting
Controlled Unclassified Information in Nonfederal Systems and Organizations (NIST SP
800-171 Rev. 2) mandates that information deemed as CUI Basic be protected in the same
manner as information with security classifications “at higher than moderate confidential-
ity” as identified in the Standards for Security Categorization of Federal Information and
Information Systems (FIPS PUB 199) and the Minimum Security Requirements for Federal
Information and Information Systems (FIPS PUB 200).97 These requirements concern over a
dozen different families of practices and procedures to protect CUI including how to control
access to CUI, how to audit information systems, and how to physically protect CUI,
among others.
Everyone with access to CUI must undergo mandatory training—whether they are inside
or outside the government—and NIST has created detailed processes for organizations
to assess their preparedness for protecting CUI and to address advanced persistent threats
(particularly cyber threats).98 NIST and DOD acquisition officials work closely with private
industry representatives to build in some flexibility for the protection methods, which
allows them to tailor their approach so that technology solutions are not overengineered.
Some CUI might require “expensive [and] exquisite capabilities” to protect against unlawful
access, but Pentagon officials have said that they “don’t want companies spending money to

buy security that [they] don’t feel will be necessary to protect the Department of Defense
information and national security information.”99
Because CUI is created by so many agencies that are granted the prerogative by different
statutes, the penalties and sanctions associated with unauthorized disclosure of CUI are
wide ranging. Violations of export controls in the dissemination of CUI research are subject
to the 2018 Export Control Reform Act. Pursuant to section 4819 of that title, the U.S.
government can impose criminal and/or civil penalties. Criminal sanctions include a fine of
not more than $1 million and/or a prison sentence of not more than twenty years per viola-
tion. Civil sanctions include a fine of not more than $300,000 per violation. The secretary
of commerce can also revoke licenses or bar individuals/entities from participating in the
United States’ export licensing regime.100
Moreover, because CUI covers such a vast breadth of information—from human resources
information to nuclear naval propulsion technologies—the process for accessing CUI varies
widely. As a rule, information that is deemed CUI does not require a security clearance
to access. However, there are cases where an individual would need a security clearance
to access the systems or facilities that maintain the CUI, thereby necessitating a security
clearance, or where the specific CUI of interest requires a security clearance and background
investigation under a particular statute or regulation. In all cases, those handling, process-
ing, storing, or maintaining CUI must have a lawful government purpose when doing so.101
It is important to note that CUI is a marking that can only be applied to information that
is created for or by the government and needs to be protected pursuant to a law, rule, or
government policy. That is, research on sensitive topics may be controlled but cannot be
deemed CUI unless the research was conducted by or for the government and/or is subject
to a government contract or agreement. Nevertheless, BIS may at some point decide that
sensitive research should be subject to export controls and subsequently be treated in a
manner similar to CUI—or even receive the CUI marking—in line with the U.S. govern-
ment export-licensing regime. This is a process of which corporate export control compliance
offices are keenly aware.102
The Japanese Information Security System

for Science and Technology
Since World War II, Japan’s treatment of national security and military secrets has been
the subject of intense scrutiny by academics, opposition parties, and the general populace.
Concerns that the pacifist and democratic legal protections established after World War
II could be undermined have dogged the dominant LDP and curtailed reform initiatives.
The LDP has managed to strengthen the national security establishment over the years, for
example by elevating the former Defense Agency to the Ministry of Defense (MOD) and
allowing the use of outer space for defensive purposes, among other moves. But often, these
reforms have exacted a high political cost or been watered down from the LDP’s original
proposals. In other cases, they have been stymied altogether, such as former prime minister
Shinzo Abe’s attempts to revise Article 9—the clause of Japan’s constitution that renounces
war—that were opposed by most Japanese citizens.103
The Abe government did make progress in 2013 with the establishment of a National
Security Council (NSC) similar in function to the U.S. body of the same name, as well
as the passage of the Act on the Protection of Specially Designated Secrets (SDS). These
deliberately paired moves were intended to strengthen the prime minister’s ability to manage
national security policy by improving the decisionmaking process and upgrading control
over sensitive information. The original proposals came under fire from a skeptical public
worried about the government’s newly proposed capacity to keep some state secrets classified
nearly indefinitely and with limited independent oversight. Although the legislation was
modified to ameliorate criticism, it officially took effect that December amid some public
protest and diminished support for the cabinet.104 The United States welcomed both moves
as important progress toward facilitating classified information sharing.
Designating Secrets in Japan
In part to assuage public worries about the scope of the government’s classification powers,
the SDS law narrowly defines the kinds of information that can be designated as national se-
crets. It enumerates four classes of information: military defense (such as defense operations,
plans, cryptology for defense, and defense equipment); diplomacy; counterintelligence (to
prevent “harmful activities” in Japan); and terrorism.105 Notably, unlike the American legal
bases for classification, the SDS law does not offer classification authority for information
related to a broadly defined category of national security, which could include economic or
technology factors. The SDS list is quite specific, and it is considered a positive list, meaning
that items not listed do not qualify.106
Moreover, the SDS law does not identify classification authority for scientific information
pertaining to national security interests outside the scope of “military defense” or weapon
secrets. So, nonmilitary work in such areas as materials science or cryptographic encryption
cannot be classified by the government. The act does protect R&D information related
to defense equipment. However, to date, this has been interpreted to cover only existing
equipment or specific weapon systems with appropriated funding. It does not apply to
scientific inquiry or R&D for technologies or products that could potentially have a military
application in the future (so-called dual-use technologies).107
As a result, the number of ministries authorized to designate state secrets is relatively small,
limited to only the Ministries of Defense; Foreign Affairs; Economy, Trade and Industry;

and Justice, as well as the National Police Agency, the Cabinet Secretariat and NSC, and a
few other related agencies.108 Japan’s MEXT, for example, does not have the authority to des-
ignate state secrets, even though it oversees the Japan Atomic Energy Agency and is Japan’s
largest government provider of S&T research grants—between fiscal years 2012 and 2019, it
was responsible for at least 50 percent of research appropriations. Moreover, those ministries
and agencies with the authority to designate secrets do so relatively rarely—only 569 times
across the entire government over the first six years since the SDS law went into effect.109
Japan’s MOD is responsible for 335 of those instances, followed by the Cabinet Secretariat at
eighty-seven and the National Police Agency at forty-three.110
Personnel Clearance Processes in Japan
The SDS law also outlines the process by which government employees and private con-
tractors receive personnel security clearances to view and handle classified information.
Japan’s Cabinet Secretariat sets the standards for this process (based on the SDS law) and
provides oversight for the sake of uniformity, but each ministry or agency is responsible for
issuing individual clearances. The process of these clearance assessments in Japan is mostly
similar to the U.S. process, with candidates filling out detailed personal information forms
and undergoing background checks for verification. Both countries use similar criteria for
determining an individual’s suitability to hold a clearance.
Each ministry’s personnel office utilizes its own records and coordinates with the National
Police Agency, the Public Security Intelligence Agency, and the Cabinet Intelligence
and Research Office to check criminal records and investigate any credible allegations
of substance abuse or other potentially compromising behavior or personal relations.111
Reassessments are necessary every five years if individuals continue to require a clearance.
Eligible government contractors can also seek clearances for as long as their work on a
specific contract requires classified access.
Still, because the SDS law places strict limits on who can classify information and what can
be classified, the number of cleared individuals is relatively small compared to the United
States. In 2019, for example, Japanese government ministries and agencies evaluated about
23,000 individuals for clearances related to the SDS law (compared to well over 1 million
comparable investigations in the United States).112 Only 320 were eligible contractors, with
the rest government employees. Just two people failed to gain a clearance (less than .01
percent), compared to an average denial rate in the United States intelligence community of
roughly 1 percent.113 By the end of 2019, just under 135,000 people in Japan were eligible to
access SDS information, of whom only 3,403 were not in government.
Moreover, because the SDS law only covers scientific or technical information that is directly
related to military equipment (not Japan’s broadly defined national security), the govern-
ment does not have authority to grant a security clearance to a civilian researcher or private
sector employee outside of that context. MEXT, for example, does not have the authority
to issue a clearance to someone under its jurisdiction. This has significant implications for
potential Japanese technical cooperation with the United States in nonmilitary fields that
U.S. policymakers deem sufficiently sensitive to be kept secret, including some classified
R&D involving certain areas of quantum computing, materials science, AI, space systems,
and other advanced technologies. Similarly, it is unlikely that the Japanese government could
issue a clearance to individuals in the private sector so that domestic companies can use
classified cybersecurity information shared by friendly governments to protect themselves
from zero-day attacks.114
Although U.S. authorities are often reluctant to share classified information or conduct
joint projects with other countries, in some situations they clearly do want to pursue this
approach. In fact, the U.S. DOD published in late 2020 its International Science and
Technology Engagement Strategy, precisely because it aims to “strengthen alliances and
attract new partners” for the DOD S&T enterprise in line with the National Defense
Strategy.115 This goes beyond collaborative work on specific weapons systems or even de-
fense-focused laboratories to include a wider range of private sector and university resources.
“Through these alliances,” the report notes, “we may unite the allied S&T community to
rise to the challenge posed by strategic competitors such as Russia and China.” Moreover, in
2021, the Senate Foreign Relations Committee approved a draft bill known as the Strategic
Competition Act that proposes establishing a “United States-Japan national security inno-
vation fund to solicit and support private sector cooperation for new technologies that could
benefit the United States and Japan’s mutual security objectives.”116
Government-Supported Research in Japan
Since 1995, Japan’s Cabinet Office has overseen the preparation of five-year basic plans for
S&T promotion, which guide policymaking and funding.117 Japan’s sixth five-year plan
approved in 2021 reflects considerable continuity from its predecessor plan, with a notable
added emphasis on increasing public funding, incorporating digital transformation, stimu-
lating S&T innovation for problem solving by fusing the natural sciences with humanities
and social science in education, and enacting “technology outflow countermeasures.”118 The
Japanese government has a variety of tools at its disposal to help achieve its goals.
As in any country, government funding for S&T research is complicated and involves many
stakeholders (see figure 3). The Council on Science, Technology and Innovation (CSTI)
(highlighted in teal) is ultimately responsible for the basic plan, with support from a secre-
tariat staffed by about one hundred people pulled in from various ministries and agencies.
Japan’s MEXT is the biggest government player, acting essentially like the custodian or
administrator for many of Japan’s most capable science and research organizations (including
the Japan Science and Technology Agency (JST), the Institute of Physical and Chemical
Research (RIKEN), the Japan Aerospace Exploration Agency (JAXA), the Japan Atomic
Energy Agency (JAEA), research universities, and more).

FIGUR
Figure E 3
3. Government of Japan’s S&T Policy/Funding Structure
(select ministries and agencies) S&T Policy/Funding Structure
Government of Japan’s
(select ministries and agencies)
Cabinet
Council for Science, Strategic

Technology and Headquarters for
Innovations (CSTI) National Space Policy
Science Council
of Japan
Ministry of Internal Ministry of Education, Ministry of Economy

Ministry of Defense
Affairs and Culture, Sports and Trade and Industry Other Ministries
(MOD)
Communications (MIC) Technology (MEXT) (METI)
National Institute of Japan Science and Acquisition, National Institute

Information and Technology Agency Technology and of Advanced
Communications (JST) Logistics Agency Industrial Science
Technology (NICT) and Technology
(AIST)
Institute of
Physical and New Energy and
Chemical Research Industrial Technology
(RIKEN) Development
Organization
(NEDO)
Japan Aerospace
Exploration Agency
(JAXA)
National Institute for

Materials Science
(NIMS)
Japan Atomic
Energy Agency
(JAEA)
The Ministry of Economy, Trade and Industry (METI) and the Ministry of International
Affairs and Communications (MIC) are becoming more engaged and influential on this
front as well. To some extent, a competition for resources and policy direction is migrating
up to CSTI and playing out at the NSC (which, in 2020, added an economic security office
to the National Security Secretariat headed by a cabinet councilor).119 As in the United
States, a key question is how to find the right balance between technonationalism and
technoglobalism in areas of scientific exploration and technology development. Governments
want to give every reasonable advantage possible to domestic industry and invention, but
they also need open markets in other countries and as large and collaborative a global in-
novation ecosystem as possible. A related question is whether enhanced cooperation among
like-minded countries in these areas could help strike that balance. If this is a likely result,
how could Japan ensure that it is a full partner in the effort? These and similar questions are
fueling internal debate over funding, program design, and where to draw the line on certain
potential dual-use technologies related to AI and QIS (among others areas).
FIGU RE 4
Figure 4. S&T
Japan: Japan: S&T Appropriations
Budget Budget Appropriations by Ministry/Agency
by Ministry/Agency
4
Y E N (T RI L L I O NS)
0
2012 2013 2014 2015 2016 2017 2018 2019
Ministry of Education, Culture, Science and Technology Ministry of Economy, Trade and Industry
Ministry of Health, Labor and Welfare Ministry of Internal Affairs and Communication
Ministry of Defense Ministry of Agriculture, Forestry and Fisheries Others
SOURCE: "Indicators of Science and Technology, 2019," Ministry of Education, Culture, Science and Technology, 196 (24-1-2),
accessed June 2021, https://www.mext.go.jp/en/publication/statistics/title03/detail03/mext_00014.html.
Source: “Indicators of Science and Technology, 2019,” Ministry of Education, Culture, Science and Technology, 196
(24–1–2) accessed June 2021, https://www.mext.go.jp/en/publication/statistics/title03/detail03/mext_00014.html.

A look at how government funding is distributed by ministry in Japan reveals that MEXT
provides by far the most awards for S&T research, followed by METI (see figure 4).120
The recent growth in the “others” category is mostly due to increased allocations to the
Ministry of the Environment. Data from the United States shows there is no simple bu-
reaucratic counterpart for MEXT—the largest funders of R&D are the U.S. DOD and the
Department of Health and Human Services. Further, the large share of funding disbursed
by the Department of Health and Human Services can be accounted for by continuous and
generous funding of the National Institutes of Health by Congress (see figure 5).121 Still, due
to strong existing relationships among various agencies and departments (such as NASA-
JAXA, NSF-JST, DOE-JAEA, DOE-METI, and U.S. national laboratories and universities
with several top universities in Japan), there is an ample foundation of bilateral collaboration
upon which to build further.
Another illuminating comparison is how research is funded and conducted in Japan as

opposed to the United States (figure 6). Although the data differ in terms of funding scale,
the pattern of funding flows reveal the similarities and differences in how each country op-
erates.122 One key difference is the relatively smaller role that Japanese government funding
F I G URE 5
Figure 5. United States: R&D Expenditures by Department/Agency
United States: R&D Expenditures by Deparmtent/Agency
2000
150
D O L L A RS ( BIL L IO N S)
100
50
2012 2013 2014 2015 2016 2017 2018 2019 2020
DOD DOE NASA HHS (NIH INCL. )
NSF USDA OTHERS
SOURCE: "Total R&D by Agency, 1976-2020," American Association for the Advancement of Science, accessed May 26, 2021,
https://www.aaas.org/programs/r-d-budget-and-policy/historical-trends-federal-rd
Source: “Total R&D by Agency, 1976–2020,” American Association for the Advancement of Science, accessed
May 26, 2021, https://www.aaas.org/programs/r-d-budget-and-policy/historical-trends-federal-rd.
FIGUR E 6
Figure 6. National
Monetary R&D Flow of R&D
Flows Funding
(USD, 2018)(from source to investigator)
United States Japan
$576 Billion $173 Billion

SO U RC E PERFORMER SOURCE P ERFORM E R
63% 73% 79% 79%
10%
22%
13%
15%
Business Enterprises Government Higher Education Private Non-profits Rest of World
SOURCE:
Source: “Gross Domestic
“Gross Domestic Expenditure
Expenditure onSector
on R-D by R-D byofSector of Performance
Performance and
and Source of Source
Funds,”of Funds,” OECD.stat,
OECD.stat, accessed accessed
July
July 13, 13, 2021,
2021, https://stats.oecd.org/Index.aspx?DataSetCode=GERD_FUNDS.
https://stats.oecd.org/Index.aspx?DataSetCode=GERD_FUNDS.
plays in the national picture. Another is how little Japanese government research funding
goes to the business sector directly, compared to the United States. In addition, international
contributions (from the rest of world) are minimal in Japan, although there is a concerted
effort underway in Japan to expand the international component of its scientific endeavors.
This presents an opportunity for allies.
Japan’s so-called Moonshot Research and Development Program is a good example of this
opportunity for engagement.123 While the Moonshot Program is still organized domestically,
this multi-goal-oriented initiative is actively recruiting international collaborators. Some of
these goals involve AI, robotics, quantum computing, and sustainable resource circulation.
The Japanese government’s new push to realize a carbon neutral society by 2050 will add
funding and perhaps a greater sense of urgency to research in this latter area. The govern-
ment’s push for digital transformation and creation of a new digital agency should be anoth-
er opportunity for closer U.S.-Japan cooperation, especially involving the private sectors.
These themes are all emphasized in the sixth basic plan, which revolves around a central
vision of a so-called Society 5.0 (which is a vision for a human-centered society where high
integration of cyberspace and physical space can promote economic development and solve
social problems). A growing number of Japanese officials believe that there are scientific and
technological components of Society 5.0 (primarily in AI, quantum computing, and data

flows) that could be exploited by China to Japan’s disadvantage, unless Tokyo takes steps to
protect certain advancements.124 They see a consensus with the United States and European
Union emerging (even if details remain murky) and are considering drafting a national
economic and technology security strategy to guide the policy response.125 The Japanese
government has already decided to bolster domestic research integrity by imposing stricter
disclosure rules for university researchers applying for public funds, so that any connections
to Chinese funders or programs can be discovered before grants are awarded.126
Finally, even though direct Japanese government funding support for private sector R&D is
relatively low in Japan, the government does support domestic industry actively and coordi-
nates closely with business leaders on regulatory and trade issues. Like other governments,
Tokyo provides subsidies and tax incentives for domestic consumption of new technologies
in the auto, communications, and energy industries.127 The government has also used public
funds to support domestic tech firms teetering on the verge of bankruptcy, such as when it
merged the display businesses of Sony, Hitachi, and Toshiba to form Japan Display in 2012.128
Japan also provides regular support to its semiconductor industry in the form of manufac-
turing grants, tax incentives, and preferred loans (as do several other countries, although
nowhere close to the scale of support that China provides its domestic semiconductor firms).129
Private Sector R&D and Export Controls
As in the United States, most domestic spending on R&D in Japan comes from the private
sector and is largely beyond government control (aside from the aforementioned subsidies
and so-called administrative guidance).130 In recent years, private sector R&D spending in
Japan has exceeded that of the government by at least a factor of four.131 From an alliance
perspective, no other country’s firms spend more on R&D in the United States than Japan,
both through their own work at U.S.-based affiliates and subsidiaries as well as in partner-
ship with other U.S. firms, universities, and even government-affiliated institutes (such as
NASA’s Ames Research Center).132
Japan’s FEFTA provides the government with various traditional tools for managing private
companies’ interactions with other countries.133 The most applicable tools for technology
security involve export controls and investment restrictions, but there are some gaps in how
the law is applied that limit coverage in important ways when considering compatibility with
the United States. Although an underlying purpose of the law is to support “the mainte-
nance of peace and security in Japan and in the international community,” it is limited to
“the minimum necessary control or coordination of foreign transactions.”134
Additionally, most circumstances that allow Japan’s cabinet to decide to impose export or
investment restrictions relate to fulfilling obligations under international treaties and the
maintenance of international peace and security (such as upholding arms control agree-
ments or restricted trade in wild flora and fauna). Among these situations are transactions
“designed to provide technology pertaining to the design, manufacture or use of specific
kinds of goods.”135 This is considered a kind of service transaction that would require METI
permission in order to transfer technology information. To date, this has been interpreted
within Japan to relate only to multilateral arms control regimes such as the Wassenaar
Arrangement and the Nuclear Suppliers Group.136 The Japanese government thus has a much
more limited scope for determining what constitutes export-controlled research than does
the U.S. government.
Moreover, cabinet-imposed restrictions technically only apply to transactions between

residents and non-residents, so a foreign national scientist living and working legally in Japan
for more than six months would be exempt from such restrictions.137 For these reasons, the
Japanese government does not have a reliable and flexible measure of legal control over the
export of research data or technology design information not clearly considered dual use for
the sake of arms control. In these cases, the government relies heavily on the private sector’s
protection of company trade secrets.
Additionally, METI permission is not required for a Japanese company to hire a foreign na-
tional to participate in Japan-based R&D for the firm, or what U.S. law considers a deemed
export, if that employee will have any access to controlled technology.138 In this situation,
the United States makes its designation based on the nationality of the prospective employee,
not merely their residence.
Japan’s FEFTA does provide the cabinet with an ability to intervene in special cases regard-
ing international financial transactions, investments, imports, or exports “when it is particu-
larly necessary in order to maintain peace and security in Japan.”139 But legal interpretation
to date in Japan has not applied this concept of “peace and security” as justification for
commercial export-controlled research. So far, it has primarily been used for the deployment
of economic sanctions and trade restrictions vis-à-vis North Korea.140
Japan has taken steps in recent years to upgrade the protection of domestically produced
technology, sometimes at the urging of the United States. After Washington moved to
strengthen its foreign direct investment rules in 2018, for example, Tokyo amended its
FEFTA in 2020 to lower the purchasing approval thresholds (from 10 percent to 1 percent
ownership of the company involved) and introduced a prior notification requirement.141 This
applied to certain sectors of the economy that could pose a potential national security threat
including critical infrastructure (such as electricity, gas, communications, or transport
infrastructure) and added new categories of data processing (including semiconductors and
data storage), software related to information processing, and telecommunications businesses
(like those related to internet use and mobile communications).
An interesting test of the new rules occurred in March 2021, when Chinese tech power-
house Tencent Holdings became a major shareholder, with a 3.65 percent stake, in Japan’s
Rakuten, an e-commerce giant that has diversified into telecommunications and fintech
businesses. The fact that the Japanese government was not given prior notification confused
U.S. officials, who assumed that this kind of transaction would be covered by the amended

law—however, purely financial investments are exempt from the notification requirement.142
Although it makes sense for Japan to welcome foreign investment that does not seek access
to nonpublic technology or does not take an activist approach in the management of the
company, the challenge lies in monitoring this situation indefinitely. The Japanese govern-
ment can force divestment ex post if it finds evidence that a foreign company like Tencent
has shifted from having a purely financial interest to being an activist shareholder in the
Japanese firm. But effectively policing this dynamic will likely require more resources for
intelligence gathering and surveillance.143 Information sharing in this area of corporate
counterintelligence by the U.S. and Japanese governments could also help.
In addition to tighter scrutiny of legitimate foreign investment, Japan has also taken steps
to bolster the private sector’s capacity to protect trade secrets and technology from foreign-
ers carrying out industrial espionage. The Japanese government amended its Industrial
Competitiveness Enhancement Act in 2018 to create a set of uniform standards for domestic
industrial security and a process to certify that Japanese firms were meeting high standards.144
U.S. officials saw this as a positive step but, unlike NISPOM in the United States, Japan’s
new certification process only covers unclassified material and companies’ proprietary
information, not the protection of classified material in the private sector.145 As such, the
2018 amendments by themselves are unlikely to be sufficient grounds for adding an ISA to
the U.S.-Japan GSOMIA.
Another area of asymmetry between the two countries is Japan’s lack of a secret or classified
patent system—or, for that matter, a classified court system that can help adjudicate civil or
criminal cases that might involve classified information. Among G20 nations, only Japan
and Mexico lack a secret patent system, although Japan is considering ways that it might be
able to introduce something to protect invention secrecy.146 Many difficult questions remain
unanswered, such as which ministry or office will be given the power to classify patents and
based on what authority. The U.S. DOD, for example, can issue clearances to a company
and/or researchers following the issuance of a patent secrecy order so that they can continue
their R&D work. Japan does not have a mechanism for doing this unless it is for a specific
military-use item.
Alliance Cooperation in Theory and Practice

When comparing the United States and Japan, it is important to note that for all of the laws,
guidelines, executive orders, regulations, and procedure manuals that help make up the U.S.
information protection apparatus, the behavior of people involved is still the most important
component. Because any sort of mistake can be professionally or criminally costly, there is a
general tendency to overclassify information and require higher-level assurances than might
seem necessary based on a strict reading of the rules. For example, even though Executive
Order 13526 specifically limits the number of people with original classification authority
to relatively few senior officials, the reality is that any Defense or State Department desk
officer with a clearance must decide regularly whether to start drafting a memo and initiate a
paper trail on the “high side” (the classified portion of their computer) or the “low side” (the
unclassified portion). There is a bottom-up process of classification that takes place every day
throughout the U.S. government and among its contractors that usually defaults to overpro-
tecting information.147
When it comes to sharing information with government officials or engineers from another
country, U.S. officials are generally more comfortable doing so if their foreign counterpart
exhibits a similar approach toward protecting classified information. There are times when
a government determination to share classified information with a partner like Japan goes
through a formal interagency process that fully evaluates the relevant rules, risks, and
potential rewards. More often, though, these judgements take place informally at various
levels throughout the alliance. As a result, decisions about sharing information with Japan
are often imprecise assessments of Japan’s need to know or what benefits the United States
would accrue, in addition to an individual’s or an office’s trust and confidence in its Japanese
counterpart.
A similar dynamic plays out in the private sector when it comes to the advice provided by
legal departments or compliance officers within a company that are unsure whether some
research that they are conducting is export controlled. Researchers or engineers of a Japanese
firm (subsidiary) in the United States who conduct AI or QIS research with U.S. partners
might want to share research results with their headquarters back in Japan. First, however,
they will consult with their legal department to see if it thinks an export license might be
required (as a firm’s proprietary research is not eligible for the fundamental research exemp-
tion). In many cases, the answer is not obvious, so the safest route usually results in restrict-
ing information sharing until a clear approval can be confirmed.
Although Japan overall is well trusted and highly valued by the U.S. government and mili-
tary, its legal structure and culture of information protection is often seen as not equivalent
to the U.S. system.148 This results in situations where less information is shared than might
otherwise be the case, or additional hurdles or requirements are sought to compensate for
this lack of confidence. For example, Japan is not part of Washington’s most trusted circle of
intelligence confidants (the so-called Five Eyes network), is not considered part of the U.S.
National Technology and Industrial Base (like Canada, the UK, and Australia are), and is
not on a short list of countries that Washington exempts from stricter scrutiny of inward
investments.149 The United States does not have an ISA to its GSOMIA with Japan as it does
with Sweden, India, and others.150
In practice, substantive challenges to information sharing remain. The United States and
Japan collaborate on a variety of highly classified military projects in the defense arena, in-
cluding the F-35 joint strike fighter and the SM3-Block IIA interceptor missile. Information
sharing, however, is kept to the bare minimum necessary to carry out the work. In the

case of the SM3 missile (often touted as a successful example of co-development), each
government funds its own nation’s contracted work on the separate components—the only
information shared relates to areas of interface between those components.151
Due to the lack of a foundational agreement on defense technology cooperation (such

as an ISA to the bilateral GSOMIA), each cooperative project requires its own negotia-
tion of a special program security agreement.152 The U.S. Defense Technology Security
Administration often decides that it must confirm individual security clearance approvals
given to Japanese contractors working on the project.153 The existence of the U.S.-Japan
GSOMIA mitigates this to some degree on simple defense exports, but not when it comes to
collaborative research or co-development projects. All of this adds time, cost, and inconve-
nience to joint projects, and it contributes to an overall lack of eagerness from U.S. officials
to conduct new co-development projects with Japan.154
Similarly, the State Department’s Defense Trade Advisory Group (DTAG), which is made
up of private sector defense trade professionals advising the Bureau of Political-Military
Affairs, has considered ways to expand the use of general licenses to reduce licensing chal�-
lenges and enhance collaboration with foreign firms and countries in international coopera-
tive programs.155 Although the DTAG actively considered including Japan as a country that
should be eligible for a general license exemption, the working group assigned to this task
eventually recommended only Five Eyes countries for this streamlined procedure.156 The
sticking point, according to some DTAG members, was Japan’s insufficient treatment of
deemed exports. Japanese rules and practices surrounding the hiring of non-Japanese em-
ployees or university researchers were not perceived sufficiently rigorous for blanket export
approvals.157
It is useful to consider the case of the United States and India concluding an ISA to their
GSOMIA in 2019, given the fact that India remains on the DOE’s sensitive countries list
and has maintained a close defense industry relationship with Russia for decades. At a
not-for-attribution roundtable organized by the Carnegie Endowment for International
Peace, current and former U.S. officials explained their view that Japan would need to make
some additional improvements to its government and industrial information security systems
and culture before it could strike an ISA deal with the United States (if Japan decided that it
wanted such an agreement). Two key items would be empowering a sole designated authority
for information and technology security (such as the Indian Ministry of Defense’s joint
secretary for defense industry production) and having a cadre of physical and information
security professionals within the government to provide support (such as India’s Central
Industrial Security Force, which is supported by the Ministry of Home Affairs and the
Intelligence Bureau).
Over the past few years, Japan’s legal mechanisms, systems for the protection of sensitive
and classified information, and lack of security culture have, at times, resulted in either an
unwillingness to engage on some potential collaborations or a very limited view of what can
be done in such research partnerships in sensitive areas (like QIS, electromagnetic threats,
directed-energy technologies, and some aspects of AI), according to officials from U.S. na-
tional laboratories.158 For example, the authors of this paper are aware of two separate groups
of Japanese experts who could not obtain the information they needed from U.S. officials
and experts during the 2018–2019 period because an adequate answer would touch on
classified material: first, national laboratory experts and representatives of the Tokyo Electric
Power Company (TEPCO), a private Japanese firm, regarding certain nuclear energy facility
security issues, and second, joint METI-DOE-national laboratory experts regarding an
electromagnetic pulse resilience plan. Thus, Japan’s lack of a legal framework and attendant
system to protect this kind of nonmilitary classified information has limited the allies’ ability
to share mutually beneficial information in a timely manner.
Interestingly, the U.S. DOE subsequently engaged with Japanese counterparts at METI
and the Ministry of Foreign Affairs (MOFA) to develop a legal framework by which DOE
officials could share information related to nuclear energy security that is classified as secret
with the Japanese government.159 The allies eventually concluded an agreement between
NNSA and MOFA in 2019, although there is “still some work required to be able to imple-
ment the agreement.”160 According to DOE officials familiar with the agreement, the U.S.
side still has some questions regarding Japan’s security clearance process for people who
would have access to this information, as well as how the information would be physically
handled (including sufficiently secure safes and doors, as well as personnel training).
On the private sector side, there are a variety of limitations when it comes to unclassified
collaborative U.S.-Japan work carried out in the United States. For example, one U.S.-based
Japanese firm working as a partner with multiple U.S. universities in 2020 on QIS ran into
difficulties sharing research results with the firm’s headquarters in Tokyo due to questions
about whether those results should be considered export-controlled research.161 The research
agreements that the company has with the universities are “typical peer-to-peer” relation-
ships (covering confidentiality, nondisclosure, and jointly owned intellectual property), with
a mutually agreed upon goal and plan for publishing more than 90 percent of the project
results. The Japanese company has a team in Japan doing some related work, but the compli-
ance office of the U.S.-based subsidiary was concerned that any sharing of experimental data
(prior to publication) might be export controlled. As a result, the research could not benefit
from sharing results in real time, and the process of seeking clarification from the U.S.
Department of Commerce was still pending after at least seven months.162 There is also an
added expense to retain legal advice to pursue that clarification, all of which has a chilling
effect on the potential for expanding such collaborative R&D initiatives.
Japanese firms operating in the United States in areas of sensitive dual-use technologies are
experiencing heightened scrutiny and added costs related to technology security compliance.
When filing patent applications in the United States (via U.S.-based subsidiaries), the U.S.-
based researchers and lawyers usually do not share details with the company headquarters in
Tokyo due to concerns that the information could be export controlled or even eventually
classified.163 One researcher explained that they fill out export-control applications “all the
time,” and they are almost always approved (although several reported situations where the

hiring of a foreign national at the U.S. subsidiary was prohibited due to deemed export
controls).164 Moreover, in one case, a Japanese firm said that a wireless spectrum product
it was working on was deemed classified by the U.S. government because of the product’s
ability to hide communications, and workers involved in that project had to obtain security
clearances.165 This was manageable because the Japanese firm had already set up this partic-
ular subsidiary as a separate legal entity and could be treated as a U.S. firm (albeit subject to
FOCI) with a special security agreement, although this limits the Japanese parent’s ability to
access some subsidiary information.
Another private sector example involves the use of CUI in a joint research project. In 2019,
a Japanese tech company aimed to join an unclassified DOE project as a so-called industry
partner to a consortium that involved two national laboratories, several university teams,
and a large U.S. utility company in the western United States. The energy utility could
offer large amounts of real-life data for the researchers to utilize in an attempt to apply AI
technology for more flexible and efficient use of the electricity grid, including maximizing
the use of renewable energy. However, because the utility’s data was generated from actual
customers, it was subject to CUI data protection requirements that the Japanese firm could
not satisfy without a costly reconfiguration of its IT system. That cost made the project
unattractive to the Japanese company, so it stepped away from the consortium.166
Options and Recommendations for

Bridging the Gaps
Although Japan and the United States have a productive S&T relationship, there are several
gaps in their systems that will frustrate closer cooperation if higher information security
walls are erected around some emerging technologies in the coming years. In order to bridge
these gaps, it will primarily be up to Japan to make improvements in how it protects infor-
mation, controls technology research, bolsters industrial security, and vets some scientists
and engineers. For its part, the U.S. government should take sufficient note of improvements
that Japan is making and planning in this arena. When evaluating these steps, Washington
can be supportive and look for equivalency in terms of the outcomes rather than insist on
equivalent processes.167
Bringing about the changes necessary to allow for the expansion of S&T collaboration
between Japan and the United States, particularly in areas involving the development and
application of sensitive or classified information and technologies, will require high-level
government interactions. It is likely that this will also require changes in the way that Japan
protects information and determines what should be classified and who should receive
clearances to handle such information. There are many existing cases of allied information
sharing and active R&D collaboration involving sensitive and even classified material (for
example, in the defense arena). However, the need to collaborate in these areas is growing
rapidly along with the recognition that much in the way of advances in S&T are dual use in
nature.
If Japan decides to enhance its ability to conduct sensitive dual-use R&D with the United
States and/or to facilitate high-end dual-use trade with the United States and a select group
of other nations as part of a technology alliance, there are two main categories of steps that
it should take. The first category includes specific legal adjustments, regulatory changes, and
physical investments that strengthen technology and information security in Japan. These
kinds of steps will help bridge the tangible gaps that exist between the allies’ information
security systems.
It is highly unlikely, however, that these steps alone will change the perception among U.S.
government officials that Japan’s culture of information security is adequate. In order to
address the gray areas of discretion that frequently affect decisions about what to share and
with whom, the Japanese government should consider steps in a second category that will
have a noticeable impact on their security culture, even if it is limited to a few public and
private research centers of excellence that can become preferred partners of choice. Though
they would also be substantive, there would be a sort of public relations dynamic associated
with this type of reform because it will be critical to call attention to these changes and
reinforce this public reputation over time.
The following is a list of options and recommendations for bridging the existing informa-
tion sharing and collaborative research gaps that exist in the U.S.-Japan relationship (by
category).
Category One: Legal and Procedural Issues
At present, one of the primary limiting factors for the participation of Japanese nationals
in U.S.-sponsored sensitive scientific research is the inability for Japanese investigators or
private sector engineers and scientists to receive security clearances independent of a national
defense program. An amendment to the SDS law could add a provision similar to the one
in U.S. Executive Order 13526 that covers “scientific, technological, or economic matters
relating to the national security.” The Japanese government could choose to let METI be the
designator in these cases (and the issuer of related clearances) or have this done at an inter-
agency level that involves MEXT, METI, MOD, MOFA, and the Japanese NSC. Another
option would be for the government to give MEXT some independent authority when the
research includes MEXT-supported projects and institutions. In this way, the United States
and Japan might be able to create an initiative similar to the U.S.-UK BARI program or
to participate in work supported by the JAIC or conducted by DOE national laboratories.
The closer that this type of work gets to defense applications, the more controversial it will
be to involve MEXT, but the proposed legal adjustment could allow greater private sector

participation as qualified contractors to MOD or METI. Commensurately, the legal pen-
alties for violations by the contractors would need to be equivalent to those for government
employees afforded clearances.
By way of the extant GSOMIA between the two countries, the United States implicitly ac-
knowledges the equivalence of Japan’s clearance process.168 As clearances are the first hurdle
to clear for Japanese investigators to engage in certain scientific collaboration, creating a
mechanism that allows them to receive such clearances is a key that opens this door. That
said, these authors’ review of the U.S.-India ISA experience and the NNSA-MOFA infor-
mation-sharing agreement revealed several areas that many U.S. officials still consider to be
substandard when it comes to Japan’s security clearance process and information security
system. In other words, if Japan wanted to pursue an ISA to its GSOMIA with the United
States today (or to fully implement the NNSA-MOFA agreement), U.S. officials would likely
seek the following steps or clarifications before endorsing Japan’s system as “substantially
equivalent.”
1. One would be clarification of the “designated security authority” in Japan. For

the sake of oversight for the SDS law, Japan did create a Cabinet Committee for
Protection and Oversight to support and manage the designation of secrets and
clearance processes across the relevant ministries.169 But this group does not seem to
be an equivalent organization to the U.S. government’s NDPC. Moreover, the roles
and responsibilities for disclosure and quality control that reside in other offices such
as the Cabinet Intelligence and Research Office and the Public Security Intelligence
Agency should be clarified. Japan’s Cabinet Office includes a variety of special sec-
retariats (tokubetsu no kikan) that are empowered to manage specific policy activities
with the participation of various ministries and agencies. It should be feasible for
Japan to create some kind of national disclosure secretariat that involves representa-
tion from the Justice Ministry, MOFA, MEXT, MOD, METI, MIC, and possibly
the National Police Agency (or National Public Safety Commission).
2. A second step might be creating a process for settling legal disputes in a classified
manner. A separate classified court system is probably unnecessary, but some kind
of domestic, legally sanctioned process should be developed to handle the occasional
case in this regard. Inevitably, if the Japanese government does move to increase the
amount of protected information and the number of citizens involved in this arena,
then instances of litigation involving classified material (or potentially classified
material) will also increase. The government should anticipate this development.
3. A third step could be establishing a cadre of security professionals within the

Japanese government similar to the GS-0080 class within the U.S. government,
whose “principal purpose [is] the development and implementation of policies,
procedures, standards, training, and methods for identifying and protecting infor-
mation, personnel, property, facilities, operations, or material from unauthorized
disclosure, misuse, theft, assault, vandalism, espionage, sabotage, or loss.”170 This
includes personnel security, physical security, information security, and industrial
security. With eleven grade positions, this class offers a career path for professionals
dedicated to government security that does not exist currently across the Japanese
bureaucracy.
The Japanese government could also extend a “scientific, technological, or economic matters
relating to the national security” justification as an amendment to FEFTA. This could
strengthen the government’s ability to control exports of technology research and the hiring
of foreign nationals (deemed exports) in select situations. Combined with the aforemen-
tioned recommendation related to the SDS law, these steps would raise the bar for industrial
security in Japan and make it easier for the United States and Japan to work out general
licensing arrangements that could facilitate the exchange of controlled goods and a wider
range of sensitive information in a timelier manner. As one former U.S. government official
mentioned, “if we want a real tech alliance, then we need a means for more unfettered access
between us.”171
These two amendments (to the SDS Law and FEFTA) would also provide a stronger legal
foundation upon which to establish a secret or classified patent process, as it would not only
create a valid reason to classify an invention, but also the ability to authorize the inventors to
continue their work.
If Japan chooses to refrain from more comprehensive approaches to upgrading information

protection, it might be possible to create highly controlled research environments on a
smaller scale. For example, Japan created an AI Research Center in 2015 under the National
Institute of Advanced Industrial Science and Technology (AIST).172 In 2020, the govern-
ment began setting up eight special research bases for QIS research (including at AIST
and RIKEN).173 Japan could take steps to upgrade the physical security at a couple of key
locations (such as AIST) and turn them into designated hubs for dual-use research in collab-
oration with the United States or other partners. In addition to physical security improve-
ments, enhanced measures for research integrity and trusted workforce development could
be implemented for joint work that is not necessarily classified but requires strong security
measures.174 Japan already has the ability to conduct a thorough vetting of potential em-
ployees (in itself a form of personnel clearances) for businesses such as day care centers and
gambling facilities, so it should be possible to establish a few centers of research excellence in
Japan that also exemplify the highest standards in physical and personnel security. As more
funds, talent, and cutting-edge work are directed toward these research hubs, they will no
doubt attract imitators within Japan and help raise the standard for information security.
Category Two: Strengthening and Projecting a New Culture of

Information Security
Whatever tangible steps Japan takes to upgrade its ability to share and protect sensitive
and classified information, it will be equally important to convince would-be international

partners that Japan is committed to enforcing its new rules and creating a culture of profes-
sional excellence. As such, some of the recommendations in this category relate to actions
mentioned in category one, but the important issue at hand is the manner in which that
action is carried out.
One key item will be strict and publicly noticeable enforcement of rules and imposition of
penalties when appropriate. All countries suffer occasional lapses in security, either through
negligence or from criminal activity, but the frequency and severity can be minimized via
tough enforcement. To take one recent example, Japan’s Nuclear Regulation Authority deter-
mined in March 2021 that TEPCO allowed malfunctioning intruder-detection equipment
to remain in use at its Kashiwazaki-Kariwa nuclear power plant for about one year, “even
though its security guards were aware the alternative measures were ineffective and, as a
result, it may have been impossible to detect intruders for more than 30 days.”175 Here is a
prime opportunity for the Japanese government to carry out a high-profile investigation and
penalize the company and/or some employees to the fullest extent that the law allows. In the
United States, for example, the FBI has in recent years stepped up enforcement of university
researchers failing to disclose ties to China and backed up their rhetoric with some high-pro-
file arrests.176
Of course, the primary goal for the government is to avoid security lapses in the first
place, so adequate and continuous training is necessary at all levels. This will require a
commitment of financial and personnel resources not only at the government level but also
at the university and private sector level. This will require patience, persistence, effective
communication, and some financial support so that universities and private companies feel
like partners in this process rather than simply dragooned into a costly and complicated
government program.
Japan’s efforts in this category would benefit from an element of marketing or public rela-
tions. For example, the U.S. Justice Department established its China Initiative in 2018 to
identify priority trade secret theft cases, develop an enforcement strategy for nontraditional
collectors of U.S. technology, and educate universities about potential threats, among
other related goals.177 They created a leadership team and publicized the initiative via press
conferences and public testimony in Congress. Then they proceeded to publicize and keep a
running tally of convictions on a dedicated website (which now lists almost 100 convictions).
Japan would probably want to avoid a similarly labeled initiative and choose its own ap-
proach suited to domestic laws and politics, but the model of such a focused, well-resourced,
and high-profile initiative dedicated to technology security in Japan is worth considering.
Japan has already begun to take steps in this direction, but some political resistance that
questions the necessity or affordability of these reforms stymies their effectiveness.
Updating the 1988 U.S.-Japan Agreement on Cooperation in R&D
in Science & Technology
Additionally, in order to promote greater synergy between public and private sector entities
and between the defense and commercial sectors, the allies should consider establishing
a bilateral public-private interdisciplinary body to work in support of high-level U.S. and
Japanese policymakers. This council could draw from top laboratories and corporations to
help policymakers leverage allied S&T collaboration in new ways and delineate clear prior-
ities amid fiscal constraints. It could be similar to the consultative, private sector–oriented
Joint High-Level Advisory Panel established (but later abandoned) by the 1988 U.S.-Japan
S&T cooperation agreement, or it could be a separate bilateral commission that proposes
joint funding initiatives to the NSCs of both countries for inclusion in annual budget
proposals.178
One possible approach was developed by the authors of this paper (see figure 7). The primary
benefit of this approach is that it puts a wider range of scientists and technology specialists in
closer and more regular contact with policymakers from both countries. Such an approach
can tighten the loop that connects national strategic policy priorities, funding choices, and
knowledge about the world’s most advanced technologies. If the allies decided to launch a
joint innovation fund as envisioned in the U.S. Innovation and Competition Act of 2021,
this advisory council could be a valuable resource for both countries.
FIGU RE 7
Figure 7. U.S.-Japan
U.S.–Japan StrategicStrategic Science Council
Science Advisory Advisory Council
Concept Concept (proposal)
(proposal)

Appendix A:
U.S. Department of Energy Sensitive
Foreign Nations Control (Attachment G
DE-AC36-08GO2830 Modification M1130),
Reproduced
Attachment G - Sensitive Foreign Nations Control
In accordance with the Clause I.92, Sensitive Foreign Nations Controls, this Attachment sets
forth the requirements the contract shall comply with under this contract. (Reference DOE
Order 142.3, or superseding directives)
Foreign National Access to DOE sites, programs, information and technologies will be
approved provided the access is needed to support the program objectives of DOE and/or of
U.S. national interests.
(a) Definitions
Assignee – A foreign national who has been approved to access a DOE site, informa-
tion, or technology for a period of ore than 30 consecutive calendar days.
Foreign National – A person born outside the jurisdiction of the United States, is a
citizen of a foreign government, and has not been naturalized under U.S. law.
45
Host – The DOE or DOE contractor employee responsible for the day-to-day activities
associated with the visit or assignment.
Indices Checks – A procedure whereby a request is made to appropriate U.S.

Government agencies to determine whether information exists on a particular foreign
national.
Legal Permanent Resident (LPR) – One who has the right to reside permanently and
work in the United States. An LPR may also be known as a permanent resident alien
or Green Card holder.
Security Plan – A security plan is required to address specific site security concerns
relating to foreign national visits or assignments.
Sensitive Countries List – A list of countries to which particular consideration is given

for policy reasons during the DOE internal review and approval process for visits
and assignments by foreign nationals. Countries may appear on the list for national
security, nuclear nonproliferation, or terrorism support reasons. Those countries follow:
Algeria
Armenia
Azerbaijan
Belarus
China (People’s Republic of China)
Cuba - Terrorist
Georgia
India
Iran - Terrorist
Iraq
Israel
Kazakhstan
North Korea (Democratic People’s Republic of) - Terrorist
Kyrgyzstan
Libya - Terrorist
Moldavia
Pakistan
Sudan - Terrorist
Syria - Terrorist
Taiwan (Republic of China)
Tajikistan
Turkmenistan
Ukraine
Uzbekistan
Sensitive Visit/Assignment – A visit/assignment will be considered sensitive if:
• Sensitive Country (Citizen or Birth)
• Sensitive Subject/Sensitive Areas
• Secured Facilities (Limited Area, Protected Area, Material Access Area or

Exclusion Area)
• Represent a company, business, organization or institute from countries

identified as sensitive.
Sensitive Country National – A foreign national who was born in, is a citizen of, or
is employed by a government, employer, institution or organization, of a sensitive
country.
Visit – Access by a foreign national for 30 calendar days or less.
(b) Prior Approvals Relating to Foreign Nationals
(1) Foreign visits and assignments pertaining to DOE programs must be in accordance
with DOE Order 142.3, or superseding directives and other DOE policies furnished in
writing to the contract. All visits and assignments must be approved in advance by the
DOE Approval Authority.
(2) Sensitive visits or assignment requests must be submitted 45 days in advance in

order to allow time for an indices check to be completed.
(3) Non-sensitive visits or assignment requests must be submitted 5 days in advance.
(c) Reports relating to Foreign Visits and Assignments
Host Report Requirements – To enable the approving official to evaluate the effectiveness
of visits and assignments, and to assist in determining the desirability of future visits
and assignments, host reports are required within 5 days of the completion of the visit
or assignment.

About the Authors
James L. Schoff is a senior fellow in the Carnegie Asia Program. His research focuses
on U.S.-Japan relations and regional engagement, Japanese technology and innovation,
and regional trade and security dynamics. One of his recent publications was “U.S.-Japan
Technology Policy Coordination: Balancing Technonationalism With a Globalized World,”
published by the Carnegie Endowment for International Peace in June 2020.
Douglas E. Rake is the president and chief executive officer of Racke Strategies &
Technologies, Inc. (RST). Rake and his company concentrate on the formulation and execu-
tion of strategies for the development and growth of international collaborations in science
and technology research and development as well as their related applications and market
entry.
Joshua Levy is a James C. Gaither Junior Fellow in the Carnegie Asia Program.
49
Acknowledgments
Although no specific organization is responsible for supporting the research behind this
working paper, we are grateful to several organizations and individuals who have supported
our expanding work on technology- and science-related issues in recent years within the
Japan Initiative at the Carnegie Endowment for International Peace. While the authors are
solely responsible for the research and policy recommendations explained in this working
paper, we have benefited from a variety of supporters along the way including from: The
Japan Foundation Center for Global Partnership, the Carnegie Endowment for International
Peace (both the Asia Program and the Technology and International Affairs Program),
JETRO, the Embassy of Japan in Washington, DC, and supporters of our U.S.-Japan
Defense Equipment Cooperation Study Group meetings that include Mitsubishi Heavy
Industries, Northrop Grumman, IHI Inc., Faegre Drinker, and Itochu Corporation.
We are also grateful to the large number of individuals who agreed to be interviewed for
this research including representatives of several U.S. and Japanese companies, government
offices, science and research institutions, and think tank scholars from both countries. The
authors would also like to thank the high-quality support and editing team at Carnegie
including Alex Taylor, Samuel Brase, Haley Clasen, and Jocelyn Soly.
51
Notes
1 Examples of existing bilateral agreements include the 1954 Mutual Defense Assistance Agreement, the 2007
GSOMIA, and a 2019 agreement between the Japanese MOFA and the U.S. DOE’s NNNSA that allows for
some sharing of classified information up to the secret level.
2 U.S. Office of Personnel Management, “Position Classification Standard for Security Administration Series,
GS-0080,” December 1987, accessed July 7, 2021, https://www.opm.gov/policy-data-oversight/classification-
qualifications/classifying-general-schedule-positions/standards/0000/gs0080.pdf.
3 This assessment of the U.S. view of Japan’s information security system derives from dozens of author
interviews with current and former U.S. government officials involved with these issues.
4 James L. Schoff, “Charting the Post-Cold War U.S.-Japan Alliance,” Carnegie Endowment for International
Peace, January 11, 2017, https://carnegieendowment.org/publications/interactive/us-japan-initiatives.
5 Based on several author interviews with U.S. and Japanese government officials, 2021.
6 Mallory Shelbourne, “SECDEF Nominee Austin Affirms Threat From China, Will ‘Update’ National
Defense Strategy,” USNI News, January 19, 2021, https://news.usni.org/2021/01/19/secdef-nominee-austin-
affirms-threat-from-china-will-update-national-defense-strategy; and Jim Garamone, “Esper Discusses
Moves Needed to Counter China’s Malign Strategy,” U.S. Department of Defense, August 27, 2020,
https://www.defense.gov/Explore/News/Article/Article/2326863/esper-discusses-moves-needed-to-
counter-chinas-malign-strategy.
7 For example, in 2020 Senators Chuck Schumer (D-NY) and Todd Young (R-IN) sponsored the “Endless
Frontier Act” that, among other measures, would have installed a technology directorate within the National
Science Foundation with “DARPA-like authorities” to oversee research and award grants. The Endless
Frontier Act has since been included in the Innovation and Competition Act. See Joe Gould, “Schumer Says
Senate Will Draft Tech Research Funding Bill,” DefenseNews, February 23, 2021, https://www.defensenews
.com/congress/2021/02/23/schumer-says-senate-will-draft-tech-research-funding-bill. See also Mitch
Ambrose, “Senate Fast-Tracks a Bill to Boost Domestic Technological Innovation,” Physics Today, February
26, 2021, https://physicstoday.scitation.org/do/10.1063/PT.6.2.20210226a/full.
8 This provision is included in the Endless Frontier Act, which was later amended and included in the
Innovation and Competition Act. Though the Innovation and Competition Act passed the Senate in June of
2021 with bipartisan support, its fate in the House of Representatives remains unclear at time of publishing.
53
Additional funding was also included for the development of semiconductor manufacturing capacity and
R&D in wireless communication technologies. See “The U.S. Innovation and Competition Act: Senate
Passes Sweeping $250 Billion Bill to Bolster Scientific Innovation and Compete With China,” Sidley, June
16, 2021, https://www.sidley.com/en/insights/newsupdates/2021/06/an-overview-of-the-united-states-
innovation-and-competition-act. See also John D. McKinnon, “China Rivalry Spurs Republicans and
Democrats to Align on Tech Spending,” Wall Street Journal, April 14, 2021, https://www.wsj.com/articles/
china-rivalry-spurs-republicans-and-democrats-to-align-on-tech-spending-11618427797.
9 “Gross Domestic Spending on R&D,” Organisation for Economic Co-operation and Development, accessed
July 7, 2021, https://data.oecd.org/rd/gross-domestic-spending-on-r-d.htm; and Bureau of Economic
Analysis, “Activities of U.S. Affiliates of Foreign Multinational Enterprises in 2018,” Survey of Current
Business 100, no 12. (December 2020): https://apps.bea.gov/scb/2020/12-december/1220-affiliates.htm.
10 Rintaro Hosokawa, “China Overtakes US as Leader in International Patent Filings,” Nikkei Asia, April 8,
2021, https://asia.nikkei.com/Business/Technology/China-overtakes-US-as-leader-in-
international-patent-filings.
11 Japanese Ministry of Foreign Affairs, “The 14th Japan-U.S. Joint High Level Committee (JHLC) Meeting
Under the Agreement Between Japan and the U.S. on Cooperation in Research and Development in Science
and Technology,” May 3, 2019, https://www.mofa.go.jp/press/release/press4e_002440.html.
12 U.S. Department of State, “Tokyo Statement on Quantum Cooperation,” December 19, 2019, https://www
.state.gov/tokyo-statement-on-quantum-cooperation. This shared priority built on a foundation detailed in
U.S. Department of Energy and Japanese Ministry of Education, Culture, Sports, Science and Technology,
“Implementing Arrangement Between the Department of Energy of the United States of America and the
Ministry of Education, Culture, Sports, Science and Technology of Japan Concerning Cooperation in the
Field of Research and Development of Innovative Nuclear Energy Technologies,” February 8, 2005, https://
www.energy.gov/sites/prod/files/4.5.1.1.3.5_japan_agreement2.pdf; High Energy Accelerator Research
Organization, “KEK and the U.S. Department of Energy (DOE) Signed a Project Arrangement Concerning
High Energy Physics,” October 9, 2015, https://www.kek.jp/en/NewsRoom/Release/20151009153000; and
Jim Kurose, “Dear Colleague Letter: National Science Foundation (NSF)—Japan Science and Technology
Agency (JST) Collaborative Research,” National Science Foundation, NSF 17-077, April 24, 2017, https://
www.nsf.gov/pubs/2017/nsf17077/nsf17077.jsp.
13 “Fact Sheet: U.S.-Japan Competitiveness and Resilience (CoRe) Partnership,” White House, April 16,
2021, accessed July 7, 2021, https://www.whitehouse.gov/briefing-room/statements-releases/2021/04/16/
fact-sheet-u-s-japan-competitiveness-and-resilience-core-partnership/#:~:text=The%20United%20States%20
and%20Japan%20have%20launched%20a%20new%20partnership,energy%20and%20other%20
relevant%20sectors.
14 Arindrajit Basu and Justin Sherman, “Two New Democratic Coalitions on 5G and AI Technologies,”
Lawfare (blog), August 6, 2020, https://www.lawfareblog.com/two-new-democratic-coalitions-
5g-and-ai-technologies.
15 For more on initiatives with military connections, see U.S. Department of Defense, Department of Defense
International Science and Technology Engagement Strategy (Washington, DC: Department of Defense,
2020), accessed May 11, 2021, https://www.cto.mil/dod-ists/. See also U.S. Department of Defense,
“Allied Prototyping Initiative” presentation by the Office of the Undersecretary of Defense (Research &
Engineering), June 16, 2020, https://ac.cto.mil/wp-content/uploads/2020/08/api_overview_06_16_2020_
cleared.pdf. For more on commercial connections, see Identification of and Review of Controls for Certain
Foundational Technologies, 15 C.F.R. 742-774, https://www.federalregister.gov/documents/2020/08/27/
2020-18910/identification-and-review-of-controls-for-certain-foundational-technologies.
16 “LDP to Call for Economic Security Promotion Law,” Japan Times, September 28, 2020, https://www
.japantimes.co.jp/news/2020/09/28/national/politics-diplomacy/ldp-economic-security-promotion-law.
17 Akira Oikawa, “Japan Tightens Rules on Tech Theft to Safeguard Research With US,” Nikkei Asia,
April 28, 2021, https://asia.nikkei.com/Business/Technology/Japan-tightens-rules-on-tech-
theft-to-safeguard-research-with-US.
18 Note that the U.S. government can only classify information that “is owned by, produced by or for, or is
under the control of the United States Government.” See Exec. Order No. 13,526, “Classified National
Security Information,” Federal Register 73 (December 29, 2009): 707–731, accessed May 10, 2021, https://
www.federalregister.gov/documents/2010/01/05/E9-31418/classified-national-security-information.
19 Ibid.
20 The United States, unlike other countries, does not have a single state secrets law, hence the necessity for an
Executive Order that creates a harmonized standard for the many different types of potentially classifiable
information. The statutory basis for placing dissemination controls on information can be found in several
laws including the Espionage Act of 1917, the Atomic Energy Act of 1954, and the Intelligence Identities
Protection Act of 1982. See ibid.
21 Executive Order 13526 is the latest in a series of roughly twenty executive orders related to classified infor-
mation management (dating back to 1940), and it effectively overrides all previous orders. It is worth noting
that parallel to the top secret, secret, and confidential standards that were developed for classifying national
security information, the Atomic Energy Act of 1954 developed a standard of safeguarding information
directly related to nuclear energy and weapons. The “restricted data” and “formerly restricted data” markings
are applied almost exclusively by the Department of Energy and can be applied to information that is also
deemed classified. Thus, a document may receive, for example, the dissemination marking “SECRET//
RESTRICTED DATA.” See Matthew Kahn, “The Law of Classified Information: A Primer,” Lawfare (blog),
June 25, 2020, https://www.lawfareblog.com/law-classified-information-primer#:~:text=What%20law%20
governs%20classified%20information,Court%20in%20United%20States%20v.&text=Executive%20
Order%2013526%2C%20the%20most,by%20President%20Obama%20on%20Dec.
22 U.S. Department of Defense, National Industrial Security Program Operating Manual (2006), 32, accessed
May 10, 2021, https://www.esd.whs.mil/portals/54/documents/dd/issuances/dodm/522022m.pdf.
23 U.S. Defense Counterintelligence and Security Agency, “Security Assurances for Cleared Individuals and
Facilities,” accessed May 18, 2021, https://www.dcsa.mil/mc/ctp/int/security.
24 Government of the United States of America and Government of Japan, “U.S. and Japan Mutual Defense
Assistance Agreement,” March 8, 1954, accessed July 7, 2021, https://people.unica.it/annamariabaldussi/
files/2015/04/USA-Japan-Treaty-1954.pdf.
25 Japanese Ministry of Foreign Affairs, “Agreement Between the Government of Japan and the Government
of the United States of America Concerning Security Measures for the Protection of Classified Military
Information,” August 10, 2007, https://www.mofa.go.jp/region/n-america/us/security/agree0708.html.
26 Article 1 of the GSOMIA states that “[CMI] means any defense-related information that is generated by
or for the use of or held by the Department of Defense of the United States of America or the Ministry of
Defense of Japan, or defense-related information generated by or for the use of or held by other relevant
authorities of the Government of the United States or the Government of Japan that requires protection
in the interests of national security of the originating party. . . . Such information may be in oral, visual,
electronic, magnetic, or documentary form, or in the form of equipment and technology.” See ibid.
27 U.S. Department of Defense, “Types of International Programs” in International Programs Security Handbook
(2009), accessed May 10, 2021, https://www.dscu.mil/pages/resources/ips.aspx.
28 U.S. Department of Defense, “National Disclosure Policy” in International Programs Security Handbook
(2009), accessed May 10, 2021, https://www.dscu.mil/documents/publications/international_programs_
security_handbook/Chapter3_04052010.pdf.
29 Ibid.
30 An ISA is a government-to-government agreement that is attached to a broader GSOMIA. It, in effect,
acknowledges the equivalency of a partner government’s information security regime for the partner nation’s
nongovernment entities. This allows for closer international cooperation and collaboration between the two
nation’s defense industries and contractors.
31 U.S. Department of Defense, “International Agreements” in International Programs Security Handbook
(2009), accessed May 10, 2021, https://www.dscu.mil/documents/publications/international_programs_
32 U.S. Department of Defense, “BARI: Bilateral Academic Research Initiative: International
Partnerships for High-Impact Science,” September 25, 2018, https://basicresearch.defense.gov/Pilots/
BARI-Bilateral-Academic-Research-Initiative.

33 U.S. Department of Defense, “Types of International Programs” in International Programs Security Handbook
(2009), accessed May 10, 2021 https://www.dscu.mil/documents/publications/international_programs_
34 U.S. Department of Defense, “Coalition Warfare Program (CWP) Frequently Asked Questions (FAQs),”
accessed May 10, 2021, https://www.acq.osd.mil/ic/cwp.HTML.
35 Ibid.
36 U.S. Department of Defense, “Allied Prototyping Initiative,” presentation by the Office of the
Undersecretary of Defense (Research & Engineering), June 16, 2020, https://ac.cto.mil/wp-content/
uploads/2020/08/api_overview_06_16_2020_cleared.pdf.
37 National Defense Authorization Act for Fiscal Year 2021, Pub. L. No. 116-283, (2020), https://www
.congress.gov/bill/116th-congress/house-bill/6395.
38 U.S. Department of Defense, “Joint Artificial Intelligence Center,” accessed July 7, 2021, https://dodcio
.defense.gov/About-DoD-CIO/Organization/jaic.
39 Author interview with a DOE National Laboratory official in charge of security, phone call, March 3, 2021.
40 National Security Council, National Policy on the Transfer of Scientific, Technical and Engineering Information,
National Security Decision Directive 189 (Washington, DC: National Security Council, 1985), https://
fas.org/irp/offdocs/nsdd/nsdd-189.htm; and U.S. Department of Energy, “Compliance With U.S. Export
Control Laws, Regulations, and Policies,” in DOE Acquisition Guide (2021), 12, https://www1.eere.energy
.gov/ztest/fy21_acquisition_guide_fy2021_v2.pdf.
41 This case-by-case review and negotiation process also generally applies to division of intellectual property and
patent licensing rights.
42 U.S. Department of Energy, Strategic Partnership Projects [Formerly Known as Work for Others (Non-Department
of Energy Funded Work)], DOE Order 481.E (Washington, DC: Department of Energy, 2018), accessed May
10, 2021, https://www.directives.doe.gov/directives-documents/400-series/0481.1-BOrder-e-chg1-ltdchg.
43 U.S. Department of Energy, “Strategic Partnership Projects Policy in the Office of Science,” accessed July 7,
2021, https://science.osti.gov/-/media/lp/pdf/work-for-others/Office_of_Science_WFO_Philosophy
.pdf?la=en&hash=AEB03EA5D1176E9E0A089C49E74700BCEB8DE558.
44 U.S. Department of Energy, DOE Cooperative Research and Development Agreements, DOE Order 483.1B
(Washington, DC: Department of Energy, 2019), accessed May 10, 2021, https://www.directives.doe.gov/
directives-documents/400-series/0483.1-BOrder-b-chg2-ltdchg.
45 Under a CRADA (and when negotiated under other structures), in line with the Bayh-Dole Act, the U.S.
government maintains “march-in” rights. These intellectual property rights are reserved for the government
to ensure that intellectual property generated with federal funding can be commercialized. If the government
determines that the CRADA partner has abandoned efforts to commercialize or sufficiently disseminate
the investigation’s resultant intellectual property (when they still own the license to that technology), the
government can compel the partner to license the intellectual property to a third party interested in
commercializing the technology at a negotiated “reasonable” royalty.
46 Mike Furey et al., The Technology Working Group Guide to Partnering with DOE’s National Laboratories
(Washington, DC: Department of Energy, 2016), accessed May 10, 2021, https://www.energy.gov/
technologytransitions/downloads/ttwg-guide-partnering-does-national-laboratories.
47 There are many examples of Japanese entities—government agencies, private firms, and universities—part-
nering with various national laboratories using these mechanisms. For instance, the University of Tokyo
and High Energy Accelerator Research Organization have NPUAs with Fermilab; RIKEN has worked with
Brookhaven National Laboratory on sharing high-performance computing resources; the Chiyoda corpo-
ration and Tatsuno North America have CRADAs with Argonne National Laboratory and the National
Renewable Energy Laboratory, respectively; and JAXA and the Japan Atomic Energy Agency have SPPs with
Lawrence Berkeley National Laboratory. See “Institutions with Active Non-Proprietary User Agreements
(NPUAs),” Fermilab, accessed July 8, 2021, https://partnerships.fnal.gov/partnering/types-partnering/
npua-list; “National Laboratories Offer Computing Time to Japanese Physicists in Wake of Earthquake,”
U.S. Department of Energy’s Brookhaven National Laboratory, Fermi National Accelerator Laboratory,
and Thomas Jefferson National Accelerator Facility, May 23, 2011, https://www.bnl.gov/newsroom/news.
php?a=111283; “H2@Scale: Enabling Affordable, Reliable, Clean, and Secure Energy Across Sectors,” U.S.
Department of Energy Office of Energy Efficiency and Renewable Energy, September 2020, https://www
.energy.gov/sites/prod/files/2020/09/f79/h2-at-scale-crada-projects-2020.pdf; and “Our Partners,” Berkley
Lab, accessed July 8, 2021, https://spo.lbl.gov/partners/?_partner_categories=government.
48 U.S. Department of State, “The 16th Japan-U.S. Joint Working-Level Committee (JWLC) Meeting on
Science and Technology Cooperation,” June 17, 2021, https://www.state.gov/the-16th-japan-u-s-joint-
working-level-committee-meeting.
49 高エネルギー加速器兼 [Japan High Energy Accelerator Research Organization], “日米科学技術協力
事業（高エネルギー物理学分野）,” [U.S.-Japan Science and Technology Cooperation Program (High
Energy Physics)], accessed May 10, 2021, https://www2.kek.jp/kokusai/us_japan.
50 Dan Brouillette, “Department of Energy International Science and Technology Engagement Policy” (official
memorandum, Washington, DC: Department of Energy, 2018), accessed May 10, 2021, https://www
.sciencemag.org/sites/default/files/DECEMBER%20DOE%20MEMO.pdf.
51 At time of writing this paper, the S&T Risk Matrix has yet to be completed and published. According to
author interview with DOE National Laboratory official in charge of security, phone call, March 3, 2021.
52 This is an entirely hypothetical scenario with respect to nationality and mentioned research programs.
53 DOE maintains a list of countries “to which particular consideration is given for policy reasons during the
DOE internal review and approval process for visits and assignments by foreign nationals . . . for national
security, nuclear nonproliferation, or terrorism support reasons.” See U.S. Department of Energy,
Unclassified Foreign National Access Program, DOE Order 142.3B (Washington, DC: Department of
Energy, 2021), accessed May 10, 2021, https://www.directives.doe.gov/directives-documents/100-
series/0142.3-BOrder-b; U.S. Department of Energy, “Attachment G—Sensitive Foreign Nations Control,”
Modification M1130 to Definitive Contract DE-AC36-08GO28308 (Washington, DC: 2017), accessed
May 10, 2021, https://www.energy.gov/sites/prod/files/2017/02/f34/Part%20VII%2C%20SECTION%20
J%20-%20List%20of%20Documents%2C%20Exhibits%2Cand%20Other%20Attachments%20%20
Attachment%20G_0.pdf; and Jeffrey Mervis and Adrian Cho, “New DOE Policies Would Block Many
Foreign Research Collaborations,” Science, February 8, 2019, https://www.sciencemag.org/news/2019/02/
new-doe-policies-would-block-many-foreign-research-collaborations.
54 Author interviews with U.S. Department of Defense and Department of Energy officials, 2020 and 2021.
55 National Science Foundation, National Science Foundation: FY2020 Agency Financial Report (Washington,
DC: 2020), 4–5, accessed May 10, 2021, https://www.nsf.gov/pubs/2021/nsf21002/toc.jsp.
56 National Science Foundation, Proposal & Award Policies & Procedure Guide (PAPPG), June 2020
(Washington, DC: 2020), 152-153, accessed May 10, 2021, https://www.nsf.gov/publications/pub_
summ.jsp?ods_key=pappg.
57 Ibid.
58 Kelvin K. Droegemeier, “Letter to the United States Research Community” (official memorandum,
Washington, DC: Executive Office of the President, Office of Science and Technology Policy, 2019),
accessed May 10, 2021, https://aerospaceamerica.aiaa.org/wp-content/uploads/2019/09/OSTP-letter.pdf.
59 JASON, Fundamental Research Security (Washington, DC: 2019), 31–32, accessed May 10, 2021, https://
www.nsf.gov/news/news_summ.jsp?cntn_id=299700.
60 National Science Foundation, “NSF Creates New Research Security Chief Position,” March 2, 2020,
https://www.nsf.gov/news/news_summ.jsp?cntn_id=300086; and Rob Portman and Tom Carper, Threats
to the U.S. Research Enterprise: China’s Talent Recruitment Plans, U.S. Senate Permanent Subcommittee
on Investigations, Committee on Homeland Security and Governmental Affairs, 47–48, November 18,
2019, https://www.hsgac.senate.gov/imo/media/doc/2019-11-18%20PSI%20Staff%20Report%20-%20
China%27s%20Talent%20Recruitment%20Plans.pdf.
61 Author interview with a National Science Foundation official, phone call, November 23, 2020.
62 Andrew Silver, “Exclusive: US National Science Foundation Reveals First Details on Foreign-Influence
Investigations,” Nature, July 7, 2020, https://www.nature.com/articles/d41586-020-02051-8.

63 Allison C. Lerner, “Management Challenges for the National Science Foundation in Fiscal Year 2021”
(official memorandum, Washington, DC: National Science Foundation Office of the Inspector General,
2020), 5, https://www.nsf.gov/oig/_pdf/NSF_Management_Challenges_FY2021.pdf.
64 The aforementioned Innovation and Competition Act of 2021, if passed by the House of Representatives,
would expand the purview and responsibilities of federal funding institutions for research integrity. For
instance, a new interagency Research Security and Integrity Information Sharing Analysis Organization
(RSI-ISAO) would be established by the director of the White House Office of Science and Technology
Policy.
65 Office of International Science and Engineering, “Award Abstract #1952302 IRES Track-1: I/O Research
for Data-Intensive Analytics and Deep Learning,” National Science Foundation, April 7, 2020, https://
www.nsf.gov/awardsearch/showAward?AWD_ID=1952302; Office of International Science & Engineering,
“Award Abstract #1952507 IRES: Track I: Exploring Biobased Plastics and Materials Through Collaborative
Research in Japan,” National Science Foundation, April 23, 2020, accessed July 6, 2021, https://www.nsf
.gov/awardsearch/showAward?AWD_ID=1952507; Office of International Science & Engineering, “Award
Abstract #1545907 PIRE: Integrated Computational Materials Engineering for Active Materials and
Interfaces in Chemical Fuel Production,” National Science Foundation, February 2, 2021, accessed July
7, 2021, https://www.nsf.gov/awardsearch/showAward?AWD_ID=1545907; and Office of International
Science & Engineering, “Award Abstract #2020128 Collaborative Research: AccelNet: Global Quantum
Leap,” National Science Foundation, October 20, 2020, accessed July 7, 2021, https://www.nsf.gov/
awardsearch/showAward?AWD_ID=2020128.
66 Author interview with a National Science Foundation official, November 23, 2020.
67 “Chairman Menendez Announces Bipartisan Comprehensive China Legislation,” United States Senate
Committee on Foreign Relations, April 8, 2021, https://www.foreign.senate.gov/press/chair/release/
chairman-menendez-announces-bipartisan-comprehensive-china-legislation.
68 The Endless Frontier Act of 2021, for example, proposes a $100 billion increase in appropriations for the
NSF over the next five years, and the Strategic Competition Act of 2021 suggests creation of a so-called
Innovation Fund focused on U.S.-Japan collaboration.
69 Invention Secrecy Act of 195, 35 U.S.C. §181-188 (2002).
70 U.S. Department of Defense, Patent Security Category Review List: May 2009 (Washington, DC: 2009),
https://fas.org/sgp/othergov/invention/pscrl-2009-red.pdf.
71 Pursuant to Title 37 of the Code of Federal Regulations, Chapter I §5.5 and the Manual of Patent
Examining Procedure, Chapter I §120. See also Scott Locke, “The Invention Secrecy Act: The USPTO as a
Gatekeeper of National Security,” IP Theory 8, no. 1 (2019): 78-79, https://www.repository.law.indiana.edu/
ipt/vol8/iss1/4.
72 Steven Aftergood, “Invention Secrecy Activity (as Reported by the Patent & Trademark Office),” Federation
of American Scientists, accessed May 10, 2021, https://fas.org/sgp/othergov/invention/stats.html.
73 Although the VEP was established in 2008, the charter and particulars of the process were only disclosed
in full after its most recent revision and publication in November 2017. U.S. National Security Agency,
“Vulnerabilities Equities Policy and Process for the United States Government” (official memorandum,
Washington, DC: Equities Review Board, 2017), https://trumpwhitehouse.archives.gov/sites/whitehouse
.gov/files/images/External%20-%20Unclassified%20VEP%20Charter%20FINAL.PDF.
74 Ibid.
75 U.S. Department of Defense, “Foreign Ownership, Control or Influence (FOCI),” accessed May 10, 2021,
https://www.dcsa.mil/mc/ctp/foci.
76 Daniel B. Pickard, “The Defense Counterintelligence and Security Agency (DCSA) and Foreign Ownership,
Control or Influence (FOCI) Handbook,” 5–7, Wiley Rein LLP, 2021, https://www.wiley.law/media/
handbook/564_Wiley-2021-DCSA-FOCI-Handbook.pdf.
77 Note that the Defense Counterintelligence and Security Agency (DCSA) is responsible for administering
security clearances and the National Industrial Security Program. It is an agency of the DOD. It is not to
be confused with the Defense Security Cooperation Agency (DSCA), which is responsible for the DOD’s
Foreign Military Sales programs and other international security cooperation mechanisms. On certain issues,
such as the export of U.S. arms, for example, both the DCSA and DSCA would be involved.
78 Pickard, “The Defense Counterintelligence and Security Agency (DCSA) and Foreign Ownership, Control
or Influence (FOCI) Handbook”, 5–7.
79 Nicole Ogrysko, “How NBIB Slashed the Security Clearance Backlog by 300,000 in Nearly a Year,”
Federal News Network, June 17, 2019, https://federalnewsnetwork.com/nbib-transfer-to-dod/2019/06/
how-nbib-slashed-the-security-clearance-backlog-by-300000-in-nearly-a-year.
80 Katie Keller, “A Trip Down the Security Clearance Backlog Memory Lane—Ask CJ,” ClearanceJobs,
December 19, 2020, https://news.clearancejobs.com/2020/12/19/a-trip-down-the-security-clearance-
backlog-memory-lane-ask-cj.
81 Joe Davidson, “Security Clearance Process Remains ‘Broken’ Despite 2016 Changes,” Washington
Post, March 19, 2018, https://www.washingtonpost.com/news/powerpost/wp/2018/03/19/
security-clearance-process-remains-broken-despite-2016-changes.
82 Jack Corrigan, “The Pentagon Has Officially Taken Over the Security Clearance Process,” Nextgov, October 1,
2019, https://www.nextgov.com/cio-briefing/2019/10/pentagon-has-officially-taken-over-security-
clearance-process/160294.
83 Exec. Order No. 13,869, “Transferring Responsibility for Background Investigations to the Department of
Defense,” Federal Register 84 (April 24, 2019): 18125–18131, accessed May 10, 2021, https://www
.federalregister.gov/documents/2019/04/29/2019-08797/transferring-responsibility-for-background-
investigations-to-the-department-of-defense.
84 “NISPPAC Minutes,” July 15, 2020, https://www.archives.gov/files/isoo/oversight-groups/nisppac/
nisppac-minutes-july-15-2020.pdf; and “About Us,” Defense Counterintelligence Security Agency,
accessed July 7, 2021, https://www.dcsa.mil/about/#:~:text=DCSA%20services%20over%20100%20
federal,million%20background%20investigations%20each%20year.
85 “History,” Defense Counterintelligence and Security Agency, accessed May 10, 2021, https://www.dcsa
.mil/about/history; and “NISPPAC Minutes,” 29, April 14, 2021, https://www.archives.gov/files/isoo/
oversight-groups/nisppac/transcript-april-14-2021.pdf.
86 Exec. Order No. 13,691, “Promoting Private Sector Cybersecurity Information Sharing,” Federal Register 80
(February 13, 2015): 9347–9353, accessed July 7, 2021, https://www.archives.gov/files/isoo/policy-docu-
ments/eo-12829-with-eo-13691-amendments.pdf; and Information Security Oversight Office, “National
Industrial Security Program Policy Advisory Committee (NISPPAC), accessed July 7, 2021, https://www
.archives.gov/isoo/oversight-groups/nisppac.
87 “National Industrial Security Program,” Defense Counterintelligence and Security Agency, accessed May 10,
2021, https://www.dcsa.mil/mc/ctp/nisp.
88 “Attachment 1, DCSA PV Industry PSI-I Info | Continuous Evaluation, National Industrial Security
Program Policy Advisory Committee (NISPPAC) Meeting Minutes July 15, 2020,” National Archives,
accessed May 10, 2021, https://www.archives.gov/files/isoo/oversight-groups/nisppac/nisppac-minutes-
july-15-2020.pdf.
89 Information Security Oversight Office, “National Industrial Security Program Policy Advisory Committee
(NISPPAC),” accessed July 7, 2021, https://www.archives.gov/isoo/oversight-groups/nisppac.
90 Exec. Order No. 13,556, “Controlled Unclassified Information,” Federal Register 75
(November 4, 2010): 68675–68677, accessed May 10, 2021, https://www.federalregister.gov/
documents/2010/11/09/2010-28360/controlled-unclassified-information.
91 The remainder of this report’s commentary on CUI will mostly concern the EAR because it is more
expansive than ITAR topics.
92 The fundamental research exemption was established by U.S. President Ronald Reagan’s administration
and by National Security Decision Directive 189 (NSDD-189). It is important to note that NSDD-189
does not carry the force of law and is subject to an internal tension. Though it was (and remains) the stated
policy of the U.S. government to ensure that “to the maximum extent possible, the products of fundamental
research remain unrestricted,” many types of government-funded research can be restricted from publication

or general dissemination. NSDD-189 requires that government funding agencies determine “whether
classification is appropriate prior to the award of a research grant, contract [. . .] and, if so, control[ ] the
research results through standard classification procedures.” It additionally requires that “no restrictions may
be placed upon the conduct or reporting of federally-funded fundamental research that has not received
national security classification.” However, NSDD-189 makes an exception in the case that restrictions be
appropriate “as provided in applicable U.S. Statutes.” Consequently, NSDD-189 still actually permits many
types of government restrictions on fundamental research including classification and export controls.
See National Security Council, National Policy on the Transfer of Scientific, Technical and Engineering
Information, National Security Decision Directive 189.
93 Scope of the Export Administration Regulations, 15 C.F.R. §735.6-10 (2020), accessed May 11, 2021,
https://www.ecfr.gov/cgi-bin/retrieveECFR?gp=1&SID=698570cb62f6546357c5e918f6ea757e&ty=HTML
&h=L&mc=true&r=PART&n=pt15.2.734.
94 Additionally, any fundamental research concerning encryption technology as classified under
ECCN 5D002 is not covered by the fundamental research exemption unless it is publicly
available and meets the standards specified in 15 CFR VII §742.15(b).
95 “CUI Category: Export Controlled Research,” National Archives, accessed May 11, 2021, https://www
.archives.gov/cui/registry/category-detail/export-controlled-research.
96 Pursuant to Title 32 CFR XX §2002.16(a)(5) and subpart (iii).
97 Under the EAR, export-controlled research generally receives the CUI Basic designation by the BIS and
is subject to the standards enumerated in NIST SP800-171R2. However, if it receives the CUI Specified
dissemination control, additional safeguards and protections may be required. See Ron Ross et al., Protecting
Controlled Unclassified Information in Nonfederal Systems and Organizations (Washington, DC: U.S.
Department of Commerce, 2021), 3–4, accessed May 11, 2021, https://csrc.nist.gov/publications/detail/
sp/800-171/rev-2/final; National Institute of Standards and Technology, Standards for Security Categorization
of Federal Information and Information Systems (Washington, DC: U.S. Department of Commerce, 2004),
accessed May 11, 2021, https://csrc.nist.gov/publications/detail/fips/199/final; and National Institute of
Standards and Technology, Minimum Security Requirements for Federal Information Systems (Washington,
DC: U.S. Department of Commerce, 2006), accessed May 11, 2021, https://csrc.nist.gov/publications/
detail/fips/200/final.
98 “CUI Training,” National Archives, accessed May 11, 2021, https://www.archives.gov/cui/training.html.
99 Sara Friedman, “NIST Releases Draft Assessment Publication for Protecting CUI of High-Value Assets,”
Inside Defense, April 28, 2021, https://insidedefense.com/insider/nist-releases-draft-assessment-
publication-protecting-cui-high-value-assets.
100 Export Control Reform Act of 2018, 60 U.S.C. §4819 (2018).
101 Devin Casey, “CUI Q2 Stakeholders Update: Slides,” (presentation, National Archives, Washington,
DC, February 13, 2020), accessed May 11, 2021, https://isoo.blogs.archives.gov/2020/02/13/
cui-q2-stakeholders-update-slides.
102 Author interview with a National Archives official, phone call, October 22, 2020.
103 “69% Oppose Change to Japanese Constitution’s War-Renouncing Article 9, Poll Shows,” Japan Times,
June 22, 2020, https://www.japantimes.co.jp/news/2020/06/22/national/japan-oppose-change-article-9-
constitution; and “Abe Laments Missing Goal of Amending Japan’s Constitution,” Japan Times¸ May 3,
2020, https://www.japantimes.co.jp/news/2020/05/03/national/politics-diplomacy/japan-shinzo-abe-
amending-constitution-2020/. See also “Abe’s Moves Toward Collective Self-Defense,” Nippon.com,
July 11, 2014, https://www.nippon.com/en/features/h00062.
104 “State Secrecy Law Takes Effect Amid Protests, Concerns Over Press Freedom,” Japan Times, December 10,
2014, https://www.japantimes.co.jp/news/2014/12/10/national/japans-secrecy-law-takes-effect-
amid-concern-arbitrary-info-withholding-lack-oversight.
105 Cabinet Secretariat, “Overview of the Act on the Protection of Specially Designated Secrets” (presentation,
Tokyo: Cabinet Intelligence Research Office 2015), accessed May 11, 2021, https://www.cas.go.jp/jp/
tokuteihimitsu/gaiyou_en.pdf.
106 Ibid.
107 Author interview with four Japanese government officials, video call, January 12, 2021.
108 日内閣府独立公文書管理監 [Cabinet Office Independent Document Management Supervisor], “特定秘
密の指定及びその解除並びに特定行政文書ファイル等の管理について独立公文書管理監等がとっ
た措置の概要に関する報告,” [Report Regarding the Designation and Cancellation of Designated Secrets
and the Management of Specific Administrative Documents], Cabinet Office, June 19, 2020, https://www8
.cao.go.jp/kenshoukansatsu/houkoku/20200619/houkoku.html.
109 Ibid.
110 These secrets only include those designated under the SDS Act of 2013, and they do not include classified
information shared with Japan by the United States under the jurisdiction of their Mutual Defense
Assistance Agreement of 1954 (and as updated in the 2007 bilateral GSOMIA). Additionally, these are not
designations of individual documents, but subjects as secrets. By the end of 2019 some 485,000 documents
were determined to contain information designated as state secrets. See Government of the United States of
America and Government of Japan, “U.S. and Japan Mutual Defense Assistance Agreement.”
111 Author interviews with Japanese government employees and contractors who have gone through the process,
Tokyo, Japan, and video call, 2020 and 2021.
112 This comparison is a rough estimate (see page 11) because Japan’s SDS evaluations do not take account of
military personnel with access to classified military information and because investigations in the United
States are not as clearly distinct between different authorizing legislation. But the estimate provides a
roughly accurate sense of scale. See内閣官房 [Cabinet Office Secretariat], “特定秘密の指定及びその解
除並びに適性評価の実施の状況に関する報告” [Report on the Status of Evaluation, Designation, and
Cancellation of Designated Secrets], 11, Cabinet Office, June 16, 2020, https://www.cas.go.jp/jp/
tokuteihimitsu/pdf/r02_0616_houkoku.pdf.
113 Security clearance review denial rates in the United States differ between agencies, with some intelligence
agencies as high as 6.5 percent, but 1 percent is a reasonable average based on publicly available information.
See Office of the Director of National Intelligence, “Fiscal Year 2016 Annual Report on Security Clearance
Determinations,” 8, accessed July 7, 2021, https://www.dni.gov/files/documents/Newsroom/FY16-Report-
Security-Clearance-Determinations-PubRelease-20171017.pdf.
114 It is certainly possible under the current system for allied governments to share this information with
qualified government counterparts (as Japan’s SDS category of “terrorism” fits naturally), but unless private
sector employees are considered qualified government contractors in Japan, it is hard to see how clearances
could be issues to private operators of critical infrastructure such as data centers and the like.
115 U.S. Department of Defense, Department of Defense International Science and Technology Engagement Strategy
(Washington, DC: Department of Defense, 2020), accessed May 11, 2021, https://www.cto.mil/dod-ists.
116 The substance of the Strategic Competition Act has since been included as a short title in the previously
mentioned Innovation and Competition Act of 2021. See Innovation and Competition Act of 2021,
S.1260, 117th, Cong. (2021).
117 Council for Science, Technology and Innovation, Report on the 5th Science and Technology Basic Plan (Tokyo:
Cabinet Office, 2015), accessed July 7, 2021, https://www8.cao.go.jp/cstp/kihonkeikaku/5basicplan_en.pdf.
118 内閣府 [Cabinet Office], “第6期科学技術・イノベーション基本計画,” [6th Basic Plan for Science,
Technology and Innovation], Cabinet Office, accessed May 11, 2021, https://www8.cao.go.jp/cstp/
kihonkeikaku/index6.html; and 政策統括官 (科学技術・イノベーション担）[Science, Technology,
and Innovation Policy Directorate]　“科学技術・イノベーション基本計画について(答申素案),”
[Science, Technology, and Innovation Basic Plan (Draft)] 28, Cabinet Office, accessed July 7, 2021, https://
www8.cao.go.jp/cstp/stmain/20210120.html.
119 Rintaro Tobita, “Tokyo Expands National Security Council to Catch Economic Risks,” Nikkei Asia, March 18,
2020, https://asia.nikkei.com/Politics/Tokyo-expands-National-Security-Council-to-catch-economic-risks.
120 Indicators of S&T in Japan (Tokyo: Ministry of Education, Culture, Sports, Science and Technology, 2019),
24-1-2, https://www.mext.go.jp/en/publication/statistics/title03/detail03/mext_00014.html.

121 “Historical Trends in Federal R&D: Total R&D by Agency,” American Association for the Advancement
of Science, accessed July 13, 2021, https://www.aaas.org/programs/r-d-budget-and-policy/
historical-trends-federal-rd.
122 “Gross Domestic Expenditure on R-D by Sector of Performance and Source of Funds,” OECD.
stat, accessed July 13, 2021, https://stats.oecd.org/Index.aspx?DataSetCode=GERD_FUNDS.
123 “Moonshot Research and Development Program,” Cabinet Office, accessed May 11, 2021, https://www8
.cao.go.jp/cstp/english/moonshot/top.html.
124 Author interviews with Japanese officials from the Cabinet Secretariat, METI, MOD, and MOFA, Tokyo,
Japan, January and February 2020; and author interviews with Japanese officials from MEXT and METI,
January 2021.
125 “LDP to Call for Economic Security Promotion Law,” September 28, 2020, https://www.japantimes.co.jp/
news/2020/09/28/national/politics-diplomacy/ldp-economic-security-promotion-law.
126 Oikawa, “Japan Tightens Rules on Tech Theft to Safeguard Research With US.”
127 “Japan to Double EV Subsidies Using Renewable Energy,” Just Auto, November 26, 2020, https://www
.just-auto.com/news/japan-to-double-ev-subsidies-using-renewable-energy_id198971.aspx.
128 “Japan Display to Receive Full Financial Support From Government Fund,” Japan Times, August 9, 2016,
https://www.japantimes.co.jp/news/2016/08/09/business/corporate-business/japan-display-receive-full-
financial-support-government-fund; and “Japan Display,” OLED-info, accessed May 11, 2021, https://
www.oled-info.com/japan-display.
129 “U.S. Needs Greater Semiconductor Manufacturing Incentives,” Semiconductor Industry Association, July
4, 2020, https://www.semiconductors.org/wp-content/uploads/2020/07/U.S.-Needs-Greater-Semiconductor-
Manufacturing-Incentives-Infographic1.pdf.
130 Administrative guidance in Japan (or gyosei shido) refers to an informal but frequent practice (especially in
the early decades of Japan’s economic development after World War II) of Japanese bureaucrats shaping
private sector behavior through consensus building and hinting at various incentives or disincentives
that the government can leverage to elicit that behavior. Administrative guidance can be used to promote
specific private sector investments, support rationalization measures within industries, and steer corporate
behavior in certain ways that the government desires. See, for example, Chalmers Johnson, MITI and the
Japanese Miracle: The Growth of Industrial Policy, 1925-1975 (Stanford University Press, 1982), chapter 7.
The increased internationalization of Japan’s modern economy makes this approach more difficult to carry
out, but elements of administrative guidance remain in practice today. See, for example, Ryo Fujikura,
“Administrative Guidance of Japanese Local Government for Air Pollution Control,” in Development
of Environmental Policy in Japan and Asian Countries, ed. Tadayoshi Terao and Kenji Otsuka (London:
Palgrave Macmillan, 2007), 90–116, https://www.researchgate.net/publication/304730081_Administrative_
Guidance_of_Japanese_Local_Government_for_Air_Pollution_Control.
131 “Gross Domestic Expenditure on R-D by Sector of Performance and Source of Funds,” OECD.stat.
132 “About Hitachi America R&D,” Hitachi, accessed May 11, 2021, https://www.hitachi.us/rd/about-us;
and “NTT Research to Work with Caltech, Cornell, Michigan, MIT, NASDA, Stanford, Swinburne,
and 1Qbit,” Ntt Research, November 13, 2019, https://ntt-research.com/ntt-research-to-work-
with-caltech-cornell-michigan-mit-nasa-stanford-swinburne-and-1qbit.
133 外国為替及び外国貿易法 [Foreign Exchange and Foreign Trade Act], 昭和24法律第228 [Act No. 228
of Showa 24], (1949), Final revision: Act No. 102 of October 21, 2005, https://www.steptoe.com/images/
content/2/4/v1/2444/4066.pdf.
134 Ibid.
135 Ibid.
136 Author interview with a METI official, email correspondence, February 17, 2021.
137 Author interview with a METI official, email correspondence, January 12, 2021; and Overview of Japan’s
Export Controls, 4th ed. (Tokyo: 2014), 13, https://www.cistec.or.jp/english/export/Overview4th.pdf.
138 Of course, some high-end technologies (such as the most sophisticated semiconductors) can be controlled
due to their potential use with certain defense applications.
139 外国為替及び外国貿易法 [Foreign Exchange and Foreign Trade Act], 昭和24法律第228 [Act No. 228
of Showa 24], (1949), Final revision: Act No. 102 of October 21, 2005, https://www.steptoe.com/images/
content/2/4/v1/2444/4066.pdf.
141 Japanese Ministry of Finance, “Cabinet Order for the Implementation of the FEFTA on Inward Foreign
Direct Investment,” Ministry of Finance, Japan, accessed May 11, 2021, https://www.mof.go.jp/english/
international_policy/fdi/20200424.htm.
142 Mariko Kodaki, “Tencent-Rakuten Deal Exposes Limits of Japan Investment Rules,” Nikkei Asia, April 20,
2021.
143 Author interview with a METI official, email correspondence, April 21, 2021.
144 Japanese Ministry of Economy, Trade and Industry, “Cabinet Decision on the Bill for the Act on Special
Measures for Productivity and the Bill for the Act of Partial Revision of the Industrial Competitiveness
Enhancement Act,” February 9, 2018, https://www.meti.go.jp/english/press/2018/0209_002.html.
145 These two paragraphs are slightly modified from their original publication in James L. Schoff, “U.S.-Japan
Technology Policy Coordination: Balancing Technonationalism with a Globalized World,” Carnegie
Endowment for International Peace, June 29, 2020, https://carnegieendowment.org/2020/06/29/u.s.-
japan-technology-policy-coordination-balancing-technonationalism-with-globalized-world-pub-82176.
147 This assessment is based on the author’s own experience in government.
148 Lucy Craft, “Japan’s State Secrets Law: Hailed by U.S., Denounced by Japanese,” NPR, December 31,
2013, https://www.npr.org/sections/parallels/2013/12/31/258655342/japans-state-secrets-
law-hailed-by-u-s-denounced-by-japanese.
149 Scarlet Kim and Paulina Perlin, “Newly Discovered NSA Documents Shed Further Light on Five Eyes
Alliance,” Lawfare (blog), March 25, 2019, https://www.lawfareblog.com/newly-disclosed-nsa-documents-
shed-further-light-five-eyes-alliance; “Defense Primer: The National Technology and Industrial Base,”
Congressional Research Service, updated February 3, 2021, https://fas.org/sgp/crs/natsec/IF11311.pdf;
and Natsuko Segawa, “Japan Excluded From New US Foreign Investment Whitelist,” Nikkei Asia,
January 25, 2020, https://asia.nikkei.com/Politics/International-relations/Japan-excluded-from-new-
US-foreign-investment-whitelist.
150 “Industrial Security Annex to Deepen Industry Collaboration Between India, US: Nisha Desai Biswal,”
Economic Times, December 28, 2019, https://economictimes.indiatimes.com/news/economy/foreign-trade/
industrial-security-annex-to-deepen-industry-collaboration-between-india-us-nisha-desai-biswal/article-
show/73004435.cms?from=mdr; and U.S. Department of State, “Treaties and Other International Acts
Series #12852, 1997,” Agreements between the United States of America and Sweden, 14.
151 U.S. Department of Defense, Exhibit R-2, RDT&E Budget Item Justification: PB 2013 Missile Defense
Agency, DOE PE 0604881C (Washington, DC: Department of Defense, 2012), https://apps.dtic.mil/
descriptivesum/Y2013/MDA/stamped/0604881C_4_PB_2013.pdf.
152 U.S. Department of Defense, “Chapter 3: Technology Transfer and Disclosure,” in Security Assistance
Management Manual (Washington, DC: Department of Defense, 2021), accessed May 11, 2021, https://
samm.dsca.mil/chapter/chapter-3#C3.2.
153 Author interviews with former U.S. Department of Defense officials, Washington, D.C., November 2019;
and author interviews with former U.S. Department of Defense Officials, video call, September 2020. For
the F-35, for example, Defense Technology Security Administration officials traveled to Japan to review
and confirm clearances for over one hundred individual Japanese contractors. See also U.S. Department of
Defense, “Chapter 3: Technology Transfer and Disclosure.”
154 Author interviews with former U.S. Department of Defense officials, video call, September 2020.

155 “Defense Trade Advisory Group (DTAG) Plenary Meeting Minutes,” September 26, 2019,
https://www.pmddtc.state.gov/sys_attachment.do?sysparm_referring_url=tear_off&view=
true&sys_id=008709191bd51090d1f1ea02f54bcb97.
156 Candace Goforth et al., “Defense Trade Advisory Group Working Group 4: Exports, Re-exports, and Foreign
Citizenship/Foreign residence [Draft]” (Washington, DC: U.S. Department of State, 2018),
accessed July 13, 2021, https://www.pmddtc.state.gov/embargoed_countries/?id=ddtc_public_portal_
news_and_events&cat=DTAG&timeframe=all.
157 Author interview with a U.S. DTAG member, video call, February 3, 2021.
158 Author interviews with various U.S. national laboratories and multiple employees, video call, 2020.
159 Author interview with U.S. Department of Energy officials involved with the development of this agree-
ment, phone call, March 16, 2021.
160 Ibid.
161 Based on author interview with executives of the Japanese firm involved, video call, November 19, 2020 and
updated by email correspondence, June 18, 2021.
162 Based on author interview with a firm executive, phone call, June 21, 2021.
163 Author interviews with researchers at U.S.-based Japanese firms, November 19, 2020; November 25, 2020;
December 2, 2020.
164 Ibid.
165 Author interview with a Japanese research engineer, video call, November 25, 2020.
166 Author interview with researchers at a Japanese firm, video call, December 2, 2020.
167 Indeed, in many instances “substantial equivalency” for agreements such as an ISA do not require equivalen-
cy of processes but equivalency of capabilities.
168 Japan’s 1954 Self Defense Forces Law outlines three levels of security classification: kimitsu (機密), the most
restricted, gokuhi (極秘), and hi (秘), the least restricted. Under the GSOMIA these classifications corre-
spond to top secret, secret, and confidential, respectively.
169 Cabinet Secretariat, “Overview of the Act on the Protection of Specially Designated Secrets” (presentation,
Tokyo: Cabinet Intelligence Research Office 2015), accessed May 11, 2021, https://www.cas.go.jp/jp/
tokuteihimitsu/gaiyou_en.pdf.
170 U.S. Office of Personnel Management, “Position Classification Standard for Security Administration Series,
GS-0080,” 2, December 1987, accessed July 7, 2021, https://www.opm.gov/policy-data-oversight/
classification-qualifications/classifying-general-schedule-positions/standards/0000/gs0080.pdf.
171 Author interview with a former U.S. Department of Defense official, video call, December 9, 2020.
172 Japanese Ministry of Economy, Trade and Industry, “Research and Development,” Ministry of Economy
Trade and Industry, accessed May 11, 2021, https://www.meti.go.jp/english/policy/economy/research_
development/index.html.
173 “Japan to Set Up Quantum Tech Research Bases to Compete Globally,” Yomiuri Shimbun, accessed through
The Japan News, October 5, 2020.
174 So-called Trusted Workforce 2.0 in the United States is an enhanced federal personnel vetting system
launched in 2020 to incorporate “continuous vetting” and improve counterintelligence activities. See
Jory Heckman, “ODNI Preview Updated Counterintelligence Strategy, Trusted Workforce 2.0 Rollout,”
Federal News Network, February 4, 2020, https://federalnewsnetwork.com/cybersecurity/2020/02/
odni-previews-updated-counterintelligence-strategy-trusted-workforce-2-0-rollout.
175 “Security Lapses Impact Kashiwazaki-Kariwa Restart,” World Nuclear News, March 17, 2021, https://
world-nuclear-news.org/Articles/Security-lapses-impact-Kashiwazaki-Kariwa-restart.
176 Josh Campbell, “FBI Arrests Researcher for NASA Who Allegedly Failed to Report Ties to China,” CNN,
May 12, 2020, https://www.cnn.com/2020/05/12/us/nasa-researcher-arrest-china/index.html.
177 U.S. Department of Justice, “Information About the Department of Justice’s China Initiative and a
Compilation of China-related Prosecutions Since 2018,” accessed July 7, 2021, https://www.justice.gov/nsd/
information-about-department-justice-s-china-initiative-and-compilation-china-related.
178 This recommendation was first presented in James L. Schoff’s “U.S.-Japan Technology Policy Coordination:
Balancing Technonationalism with a Globalized World.” The Japan-U.S. Strategic Science and Technology
Innovation Council proposed here stems from an idea explained to the author by Douglas Rake in 2017 for
a bilateral nonprofit initiative exploring ways to derive greater benefit from current U.S.-Japan S&T collab-
oration in support of long-term shared strategic priorities. Rake, Schoff, and several others have collaborated
to realize this concept known as the Japan-U.S. Strategic Science, Technology, and Innovation Initiative
(JUSSTII).

Carnegie Endowment for
International Peace
The Carnegie Endowment for International Peace is a unique global network of policy
research centers in Russia, China, Europe, the Middle East, India, and the United States.
Our mission, dating back more than a century, is to advance peace through analysis and
development of fresh policy ideas and direct engagement and collaboration with decision-
makers in government, business, and civil society. Working together, our centers bring the
inestimable benefit of multiple national viewpoints to bilateral, regional, and global issues.
Asia Program
The Carnegie Asia Program in Washington and Beijing studies disruptive security, gover-
nance, and technological risks that threaten peace and growth in the Asia Pacific region.
67
CarnegieEndowment.org

Schoff Etall U.S.-japan Final

Uploaded by

Copyright:

Available Formats

Schoff Etall U.S.-japan Final

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Schoff Etall U.S.-japan Final

Uploaded by

Copyright:

Available Formats

WORKING PAPER

Carnegie Endowment for International Peace

This publication can be downloaded at no cost at CarnegieEndowment.org.

Introduction and Purpose 5

The U.S. Information Security System for

The Japanese Information Security System for

Alliance Cooperation in Theory and Practice 34

Options and Recommendations for Bridging the Gaps 38

Appendix A: U.S. Department of Energy Sensitive

Carnegie Endowment for International Peace 67

• A primary limiting factor for the participation of Japanese nationals in U.S.-

- clarifying its designated security authority in the government (possibly by

- creating a process for settling legal disputes in a classified manner, and

- developing a cadre of security professionals within the Japanese government

• Of course, all of this must ultimately be planned and implemented by Japanese

James L. Schoff, Douglas Rake, and Joshua Levy | 3

This paper’s objectives include:

• identifying areas of poor research protection compatibility and examples of missed

• recommending policy adjustments that can facilitate U.S.-Japan collaboration in the

James L. Schoff, Douglas Rake, and Joshua Levy | 7

Because a considerable share of U.S. R&D on cutting-edge technologies is funded by the

Executive Order 13526 also explicitly identifies “scientific, technological, or economic

compelling reasons may exist to grant access to classified information to

2. It must be consistent with U.S. military and security objectives.

James L. Schoff, Douglas Rake, and Joshua Levy | 9

Department of Defense Programs

Department of Energy Programs

James L. Schoff, Douglas Rake, and Joshua Levy | 11

• The project must be consistent with or complement the identified laboratory’s

James L. Schoff, Douglas Rake, and Joshua Levy | 13

• establishment of the bilateral Civil Nuclear Energy Research and Development

• the 2013 DOE-MEXT Implementing Arrangement Concerning Cooperation in

• a 2015 DOE-MEXT Project Arrangement concerning high energy physics;49

• a 2015 Implementing Arrangement Between the Department of Energy and the

• a 2019 DOE-MEXT Project Arrangement concerning high energy density science.

Government-Supported Unclassified Science and Technology Research

James L. Schoff, Douglas Rake, and Joshua Levy | 15

Private-Sector Considerations and Classified Patents

James L. Schoff, Douglas Rake, and Joshua Levy | 17

2018 R& D B U DGE T (B I L L I O NS O F U . S. DO L L ARS)

To balance these competing interests, the government established the Vulnerabilities

James L. Schoff, Douglas Rake, and Joshua Levy | 19

Personnel Clearance Processes in the United States

Controlled Unclassified Information

James L. Schoff, Douglas Rake, and Joshua Levy | 21

1. The principal investigator(s) must intend to publish the results of fundamental

In order to ease the process of arriving at an agreement, international research partners

James L. Schoff, Douglas Rake, and Joshua Levy | 23

The Japanese Information Security System

Designating Secrets in Japan

James L. Schoff, Douglas Rake, and Joshua Levy | 25

Personnel Clearance Processes in Japan

Government-Supported Research in Japan

James L. Schoff, Douglas Rake, and Joshua Levy | 27

Council for Science, Strategic

Ministry of Internal Ministry of Education, Ministry of Economy

National Institute of Japan Science and Acquisition, National Institute

National Institute for

Ministry of Defense Ministry of Agriculture, Forestry and Fisheries Others

James L. Schoff, Douglas Rake, and Joshua Levy | 29