Scribd
Upload
21 views3 pages

Google Safer With Google Safety API - Two-Pager R12 Opt-03

Uploaded by

Andra Tataru
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
21 views3 pages

Google Safer With Google Safety API - Two-Pager R12 Opt-03

Uploaded by

Andra Tataru
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

To protect users and other organizations from bad actors, we build and share Application

Programming Interfaces (APIs) that detect and block online threats, and help make the
internet safer for everyone.

BUILD
We create safety APIs to protect across three key
areas. Some examples of these APIs are:
Child Safety Security Information Quality
• Content Safety API • Safe Browsing API • Perspective API
• CSAI Match API • Project Shield API • Vision API
• Hash Matching API • VirusTotal AP • FactCheck API
• Civics Information API
• Text Moderation

SCALE
To scale our impact we share our safety APIs with partners:
46M 2B 4B
In 2022, Cloud Armor API Perspective API is called by The Child Safety Toolkit
blocked one of the largest 1000+ partners almost 2B (Content Safety API + CSAI
Layer 7 DDoS attacks in history, times a day. Match API) processed 4B+
at 46M requests per second. images and videos in the
last 30 days.

LIMIT
To ensure only trusted partners have access to our
APIs and that they are used in a compliant and safe
manner, we take three critical steps:

Vet Restrict Enforce

We limit access of our safety APIs to carefully vetted partners, restrict unauthorized usage of
these tools, and strictly enforce our terms of service policies. Empowering our trusted partners
with safety APIs allows us to work together, and create a safer internet for everyone.

safety.google
Google’s Safety APIs
Child Safety APIs

Child Safety Toolkit

Content Safety CSAI Match Hash Matching


Google’s proprietary model YouTube’s proprietary We provide Google’s Hash
helps partners classify technology (and the first Matching API to NCMEC to
and prioritize billions technology to use hash- help them prioritize and review
of Child Sexual Abuse matching) to identify CyberTipline reports more
Material images. It uses ML Child Sexual Abuse efficiently, allowing them
classifiers to help identify Imagery (CSAI) videos to hone in on those reports
never-before-seen CSAM and systematically flag for involving children who need
imagery for fast review, partners to review, confirm, immediate help.
removal, and reporting. report, and act upon.

Security APIs

Google’s WAAP Solution

Cloud Armor reCAPTCHA Apigee


Protects websites, services, Defends websites from Google Cloud’s full lifecycle API
and APIs against advanced fraudulent activity, spam, and management platform enables
targeted and automated other abuses. reCAPTCHA businesses to design, secure,
Distributed Denial of Service Enterprise uses an adaptive deploy, monitor, and scale APIs.
(DDoS) attacks, which risk analysis engine to keep
censor information. automated software from
engaging in abusive activities.

Safe Browsing VirusTotal


Our Safe Browsing API Allows developers to
allows client applications submit files or URLs for
to check URLs against our analysis and receive a
constantly-updated lists report on whether or
of unsafe web resources. not they are infected
This tool protects 5 billion with malware.
devices a day by warning
users against sites that
host malware or
unwanted software.

safety.google
Information Quality APIs

Perspective Vision FactCheck


Open-source API that uses Allows developers to easily Aims to facilitate the work of
machine learning to identify integrate vision detection fact checkers, journalists and
“toxic” comments. Used not features within their researchers in discovering
only to facilitate interactions applications. This includes what has & hasn’t been
between humans, but also image labeling, face and debunked globally. With this
to help promote high-quality landmark detection, optical tool, users can search claims
human interactions with character recognition from among 300K fact checks
large language models and (OCR), and tagging of from reputable publishers.
generative AI. explicit content.

Civic Information Text Moderation


The Civic Information API Google’s Text Moderation
allows developers to build tool, available through the
applications that let citizens Cloud Natural Language API,
and voters in the U.S. helps organizations scan for
know about their political sensitive or harmful content.
representation, ballot This tool is able to identify
information, and voting a wide range of harmful
locations. During supported content, including hate
elections, voters can look up speech, bullying, and
their polling place, early vote sexual harassment.
and dropbox locations, and
candidate information.

Every day We invest in building and sharing APIs that detect and block
you’re safer online threats to ensure that every day you’re safer with
with Google Google. Visit our Safety Center to learn more about how we
keep more people safer online than anyone else.

safety.google

You might also like