See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/367332544

Artificial Intelligence in Cloud Computing Security

Article · January 2023

CITATIONS READS
0 214

3 authors:

Zartashya Asharaf Avinash Ganne

University of Gujrat Raleys
82 PUBLICATIONS 0 CITATIONS 10 PUBLICATIONS 36 CITATIONS

SEE PROFILE SEE PROFILE

Noman Mazher
University of Gujrat
50 PUBLICATIONS 226 CITATIONS

SEE PROFILE

All content following this page was uploaded by Noman Mazher on 22 January 2023.

The user has requested enhancement of the downloaded file.

 Artificial Intelligence in Cloud Computing
Security
Zartashya Asharaf1, Mr. Avinash Ganne2,Noman Mazher3

Univresity of Gujrat,Pakistan1,2

Sr. SAP Basis Cloud Architect, Raley’s, Sacramento, California, USA3.

ABSTRACT
Cloud computing (CC) provides users with online access to network services,
including enhanced, transparent user management and the capacity to gather and
process data. A shared Internet gateway is offered by CC, which is evolving into a
private and public data center set. The integration of AI technologies into a
substantial amount of computer resources, especially integrated systems, presents a
number of resource problems that require careful adjustment. The IoT paradigm
recently evolved in an application for smart environments. Security and privacy are
seen as being of the utmost importance in any smart IoT environment in the real
world. Digital environments face security threats as a result of security flaws in IoT-
based networks. Because AI has excellent learning capabilities, it is more dependable
and effective at spotting harmful threats. The current architecture presented in this
article will support several applications of AI in digital homes detailed study of
security concerns and issues.

I. INTRODUCTION
Since the first invention of the client-server paradigm in 1958, cloud computing has
been driven by creative communications and distributed architecture. Rapid growth
in cloud infrastructure has made it a necessary tool for many parts of society,
including academic institutions, governmental organizations, and commercial
enterprises. Furthermore, contemporary incarnations of technology, such as
serverless computing, enable the autonomous deployment of innovative energy-use
patterns. The customizable virtual machine based on containers will boost cloud use
and offer low latency for the database environment. Deep learning techniques based
on Artificial Intelligence (AI) are projected to be used in cloud computing to predict
regional resource demand as well as new equipment architecture and planning
guidelines[1, 2]. The goal of AI as technology is to create tools that need information,
whether that knowledge comes from human understanding or knowledge gleaned
from encounters and forecasts in the past[3]. Additionally, it is anticipated that AI-
based safety tactics would be more effective at responding to new dangers than
traditional safety measures[4]. Deep learning, classification algorithms, and other AI-
based techniques have lately been suggested as effective ways to address security
concerns[5]. AI-based solutions have become more prevalent in business and other
applications as processing power and data availability increase[6-8]. A wealth of
information is accessible to prevent risks by analyzing and identifying trends of
susceptibility utilizing AI approaches. In order to assess, detect, and mitigate security
issues, each IoT device should consider having IoT-based data capabilities. It is now
considerably more reliable to identify potential malware assaults from a large
volume of data. AI is also well adapted for identifying and countering emerging and
ongoing dangers, such as those that may go unnoticed for an extended period of
time. It is crucial to develop IoT regulations and choose various factors due to the
expanding IoT and such clever assaults for potential interaction in various and
complex networks in the security protocols.[9]

II. Cloud computing

Cloud computing is the use of the Internet to access various services. In addition to
data storage, this includes hardware and systems for servers, databases, networking,
and applications[10-13]. Files may now be stored in a central archive rather than on
a personal hard drive or storage device, thanks to cloud computing [10]. Any time a
digital user gets access to the Internet, they may access apps and information.
People and businesses alike embrace cloud storage because it offers financial
savings, increased productivity, speed, dependability, and customizability.[14]

Cloud Security Threat

CC is a phrase that will see significant and wide development. There are several
threats to difficulties with protection and security. The CIA feature and cloud risks
are the basis for the analysis. Confidentiality, integrity, authenticity, and availability
have been identified by CC as the main weaknesses. Below, we briefly discuss these
issues.

Figure.1 Security threats in Cloud computing

Threats to confidentiality include issues with the software, risk from external
dangers, and threats to client information from inside. An important security risk to
client data is the illegal or unauthorized use of personal information by an intruder of
a cloud service provider. Second, the likelihood of outside assaults is quite high for
cloud systems in exposed places. This approach incorporates centralized hardware or
software for online and cloud services. Third, data loss is a vulnerability that cannot
be avoided when negotiating cloud-related terms. It is caused by human error, a lack
of tools, and, whenever feasible, access failures.

Threats to integrity include those related to record division, lax client access controls,
and data level vulnerability. First, there is a chance that customers' off-base virtual
servers and poorly educated VM design will result in exclusion from facts that are
vaguely related to protection requirements[15, 16]. Customer connections are
provided by this sophisticated cloud challenge; asset adjustments might affect the
data's dependability. The second issue is bad client access management, which has a
variety of issues and dangers and enables attackers to harm data assets owing to
improper influence and personality sharing.[17]

The equipment and its associated components are original and reliable thanks to the
authentication process. For instance, different treatment facilities within the medical
and health system receive patient criteria. The patient's therapy will be difficult if
someone else handles and obtains this information.[18]

Risks to transparency, panel expansion's effects, organizational inaccessibility,

external equipment disruption, and insufficient recovery methods are some of the
threats to availability. First, attention has been drawn to the board, which also
includes the results of basic adjustments and the implications of different
consumers' entrance into test customers. Infrastructure, as well as cloud condition
impacts, have a negative impact on how usable cloud companies are. The second
issue is the inability to access systems, which includes DNS applications, properties,
and device data transmission. a threat from outside that is present in all cloud
iterations. Thirdly, there is a physical disruption to an institution that specializes in
big networks (WAN), cloud users, and providers of IT services. Fourth, inadequate
recovery mechanisms, such as ineffective failure recovery, impact retrieval time and
effectiveness if a stage occurs.[19]

Criteria for IoT applications' security

The development of several technologies, including those used in hospitals, smart
grid applications, smart cities, smart homes, etc., is made possible by the Internet of
Things. The emergence of limited IoT and IoT systems causes new security and
confidentiality issues in these crucial applications. In this section, we discuss a few
pertinent IoT implementations and illustrate the safety issues and specifications of
each application.

Smart Grids
A key contributor to economic growth and a commodity with ever-increasing
commercial importance is electricity. Modern IT techniques are being used to
maximize electricity production while taking customer demand along the power
distribution line into account. The smart grid is the main emphasis of this
transmission line. It consists of a linked network connecting power-producing
facilities and end users, commonly known as the smart metering system (AMI),
whose primary purpose is to coordinate energy generation with end-user demand.
[20]

Several studies emphasized the necessity for intelligent grids to satisfy safety
requirements. The most important security and protection criteria are highlighted by
users. For requests for automation and control orders, network infrastructure, smart
meters, and control centers should always be available. Additionally, authorized
users are not rejected by unauthorized users when processing inquiries. Smart
meters and control systems exchange sensitive information and requests, which
cannot be disclosed to uninvited parties. They are highly useful for making choices
about the information communicated by the intelligent system in terms of
maximizing energy transfer. It is crucial for making sensible judgments that this data
be reliable. [21]

Healthcare
In order to sense and record actions, sensors and actuators have been incorporated
into patient bodies for use in healthcare applications. In order to monitor patient
health, IoT is employed in healthcare. The sensors that are already within the device
will effectively collect data from the patient's body and send it to the doctor. While
maintaining continual touch with the clinician, this technology has the power to
remove the patient from the main hospital network[22].

We provide the following summary of the healthcare safety standards based on

conceptual studies:

By using PHRs (Personal Health Records), which can only be viewed by medical
professionals, each particular patient may be protected from prying eyes. Safe
communication between patients and hospitals must be used to ensure the
confidentiality and integrity of the exchanged data.

Smart cities
Smart cities are one of the most important IoT applications. Although the phrase
"smart city" is not technically defined, it is a relatively recent idea that aims to
encourage the use of public resources and raise the level of service provided to
inhabitants. In this sense, sensors are employed on all roads, structures, and
intelligent cars to manage traffic, react to the environment, track sunlight-directed
illumination, prevent home accidents with alerts, etc. Data confidentiality and
sensitive data authentication are essential components of the security that smart
cities require. Authentication of users and the sources of information. Data privacy is
equally important since this data bundle responds, aids in decision-making, and
improves people's everyday lives in intelligent cities.

Setting Security Features

A new IoT device needs this procedure before it can connect to the intelligent home
system. The IoT platform provides a number of cryptographic algorithms that are
adapted to the applications and systems that make use of the device's
confidentiality, integrity, and encryption requirements. The reliability of their
potential customers is essential to the effectiveness and widespread acceptance of
reliable IoT infrastructures and the numerous applications they support. Such
assurance is essential due to the harm that stolen or misused private information
may do to people's social, financial, and physical life. To handle potential safety
hazards, it is crucial to make sure that proper security is implemented. This action is
crucial in defending the smart home against numerous security threats. Forged or
tainted A data-damaging incident will prevent the surveillance system from
operating as intended since judgments made using false information won't achieve
the system's intended goals, such as lowering energy consumption. The fundamental
reasons for this assault are as follows:

The intentional, suspicious setting of an IoT system produces inaccurate data. For
instance, the overall amount of power used is rising. IoT computers that process data
in-depth and fraudulently. Including electricity use as a likelihood variable, for
instance. Multi-Hop Interaction with Data Change via a Hacked IoT Gateway. IP
spoofing and identity espionage: IoT devices that connect to the same network
without encrypting the source of the data via a complex protocol, like IP sec, will be
effectively threatened by identity espionage and IP spoofing assaults.

Insights of Data on the Cloud

The cloud typically stores, processes, and analyses the data collected and
transmitted by IoT gateways. The collected data will first be archived due to the
planned use to make sure that it does not contain information on the impacted
people's attribute disclosure. There won't be any encryption of the data. Google
employs a range of data security techniques for this goal, including Data
Generalization and Differential Privacy.[23] The key problem in developing solutions
for data generalization is figuring out how to anonymize the data while limiting
identity losses arising from changes in the initial data at the same time. In order to
establish a realistic understanding of the controlled environment, important
information from examined and assessed data is then extracted and provided to the
AI-based data analytics tool, often using a machine-learning algorithm. This
information will motivate many optimal behaviors or actuator system changes. After
that, the investigated data would be preserved, and historical data would be used to
improve the development and testing of machine-learning models, modernizing the
data analytics platform to increase precision.[24]

III. CONCLUSION
Companies are producing a wide range of resources by utilizing IoT data and
computational software. These tools offer information mining solutions through the
application of statistical modeling, prediction, and classification technologies. IoT
changes how politicians make decisions. With the development of IoT and related
technologies like cloud computing, data sources may be removed from a variety of
areas. Current systems would benefit from the emergence of the IoT and AI.
Combining computerization and comprehensive analysis, development's benefits are
harvested while producing enormous economic gain. The potential to use IoT and
artificial intelligence now seem to be greater. The most difficult issues in CC are
examined in this study as security vulnerabilities.
References:
[1] T. H. Aldhyani and H. Alkahtani, "Artificial Intelligence Algorithm-Based Economic Denial of
Sustainability Attack Detection Systems: Cloud Computing Environments," Sensors, vol. 22, no.
13, p. 4685, 2022.
[2] J. P. Wahle, T. Ruas, N. Meuschke, and B. Gipp, "Are neural language models good plagiarists? A
benchmark for neural paraphrase detection," in 2021 ACM/IEEE Joint Conference on Digital
Libraries (JCDL), 2021: IEEE, pp. 226-229.
[3] J. P. Wahle, T. Ruas, N. Meuschke, and B. Gipp, "Incorporating Word Sense Disambiguation in
Neural Language Models," arXiv preprint arXiv:2106.07967, 2021.
[4] X. Fang, N. Koceja, J. Zhan, G. Dozier, and D. Dipankar, "An artificial immune system for phishing
detection," in Evolutionary Computation (CEC), 2012 IEEE Congress on, 2012: IEEE, pp. 1-7.
[5] J. P. Wahle, T. Ruas, S. M. Mohammad, and B. Gipp, "D3: A Massive Dataset of Scholarly
Metadata for Analyzing the State of Computer Science Research," arXiv preprint
arXiv:2204.13384, 2022.
[6] J. P. Wahle, T. Ruas, F. Kirstein, and B. Gipp, "How Large Language Models are Transforming
Machine-Paraphrased Plagiarism," arXiv preprint arXiv:2210.03568, 2022.
[7] F. Kirstein, J. P. Wahle, T. Ruas, and B. Gipp, "Analyzing Multi-Task Learning for Abstractive Text
Summarization," arXiv preprint arXiv:2210.14606, 2022.
[8] T. Ruas, J. P. Wahle, L. Küll, S. M. Mohammad, and B. Gipp, "CS-Insights: A System for Analyzing
Computer Science Research," arXiv preprint arXiv:2210.06878, 2022.
[9] M. M. A. Zahra, M. J. Mohsin, and L. A. Abdul-Rahaim, "Artificial intelligent smart home
automation with secured camera management-based GSM, cloud computing, and arduino,"
Periodicals of Engineering and Natural Sciences (PEN), vol. 8, no. 4, pp. 2160-2168, 2020.
[10] A. Ganne, "Cloud Computing And Security Model-A Brief Survey," International Research Journal
of Modernization in Engineering Technology …, vol. 4, no. 11, 2022.
[11] A. Ganne, "Cloud Data Security Methods: Kubernetes vs Docker Swarm," International Research
Journal of Modernization in Engineering Technology, vol. 4, no. 11, 2022.
[12] A. Ganne, "Emerging Business Trends in Cloud Computing," International Research Journal of
Modernization in Engineering Technology, vol. 4, no. 12, 2022.
[13] A. Ganne, "Applying Azure To Automate Dev Ops For Small ML Smart Sensors," International
Research Journal of Modernization in Engineering Technology, vol. 4, no. 12, 2022.
[14] Z. Xu, W. Liu, J. Huang, C. Yang, J. Lu, and H. Tan, "Artificial intelligence for securing IoT services
in edge computing: a survey," Security and communication networks, vol. 2020, 2020.
[15] R. S. Chunduri and N. Mazher, "Sas Viya 4.0 Deployment in Cloud."
[16] R. S. Chunduri and N. Mazher, "SEIBEL IP. 22X DEPLOYMENT IN CLOUD SYSTEMS."
[17] V. Kanimozhi and T. P. Jacob, "Artificial Intelligence outflanks all other machine learning
classifiers in Network Intrusion Detection System on the realistic cyber dataset CSE-CIC-IDS2018
using cloud computing," ICT Express, vol. 7, no. 3, pp. 366-370, 2021.
[18] A. Grusho, M. Zabezhailo, A. Zatsarinnyi, and V. Piskovskii, "On some artificial intelligence
methods and technologies for cloud-computing protection," Automatic Documentation and
Mathematical Linguistics, vol. 51, no. 2, pp. 62-74, 2017.
 [19] U. F. Mustapha, A. W. Alhassan, D. N. Jiang, and G. L. Li, "Sustainable aquaculture development:
a review on the roles of cloud computing, internet of things and artificial intelligence (CIA),"
Reviews in Aquaculture, vol. 13, no. 4, pp. 2076-2091, 2021.
[20] P. Radanliev et al., "Cyber risk at the edge: current and future trends on cyber risk analytics and
artificial intelligence in the industrial internet of things and industry 4.0 supply chains,"
Cybersecurity, vol. 3, no. 1, pp. 1-21, 2020.
[21] S. Sood, S. Mehmi, and S. Dogra, "Artificial intelligence for designing user profiling system for
cloud computing security: Experiment," in 2015 International Conference on Advances in
Computer Engineering and Applications, 2015: IEEE, pp. 51-58.
[22] J. P. Wahle, N. Ashok, T. Ruas, N. Meuschke, T. Ghosal, and B. Gipp, "Testing the generalization
of neural language models for COVID-19 misinformation detection," in International Conference
on Information, 2022: Springer, pp. 381-392.
[23] H. Susanto, L. F. Yie, D. Rosiyadi, A. I. Basuki, and D. Setiana, "Data security for connected
governments and organisations: Managing automation and artificial intelligence," in Web 2.0
and cloud technologies for implementing connected government: IGI Global, 2021, pp. 229-251.
[24] U. A. Butt et al., "A review of machine learning algorithms for cloud computing security,"
Electronics, vol. 9, no. 9, p. 1379, 2020.

View publication stats