AI and Future Prospects of Privacy

- Vedant Choudhary

Symbiosis International (Deemed) University

ABSTRACT

The basic definition of privacy is possessing the authority to limit, distance or seclude oneself, or
data about oneself or to restrict the influence of others on one’s life and with the popping up of
increasingly sophisticated artificial intelligence systems, the privacy concerns have only been
exacerbated. What AI brings to the fore is the capability to collect, interpret, and analyze huge
amounts of data from varied sources, thus increasing data-collection abilities of factors of society
who use this technology. AI can perform the specialized and unique tasks without any
supervision, which immensely improves the interpretation efficiency. These characteristics of AI
enable it to affect privacy in a number of different ways. Though artificial intelligence is
perceived as great advancement in majority sectors but upon critically reflecting through various
countries and places, one thing emerges: improvements and developments in technology and
science are in large part been conceived as potential threat or danger to databases and thereby
privacy and has induced the reformers and law makers, and clients to ask for more security and
protection to their data within the framework. This ideology on AI and subsequent technology as
a danger to the data privacy is also shows the thinking and ideologies policymakers. This
perspective on technology as a challenge to existing notions of, and safeguards for, information
privacy is also reflective of the mindset of contemporary law and policymaking though in areas
such as intellectual property and antitrust, law seeks to engage with technology in a more
nuanced way by enabling or in some cases leveling desired innovative or disruptive activity. This
paper seeks to answer many questions as to what should be the stance and position of law in
future to accommodate the existence of AI and how to contribute in optimizing its utility. The
future prospects of privacy and its norms would also be analyzed to bring forth a substantive
narrative of the dynamic world. The artificial intelligence aspects are critically analyzed with its
application and impact is thoroughly researched and studied.
Keywords- Artificial Intelligence, Privacy Concerns, Privacy prospects, Privacy Law,
Information Technology

INTRODUCTION

In today’s day age where life’s changing dynamically with rapid innovations and developments
taking place all around, the human species is spiraled with technologies which are the means of
their lives. Technology has become an inevitable part of human lives. Each sector has had its
share of evolution with the constant changing technologies. The artificial intelligence is the
evident and most impactful branch of the tech era. It has definitely benefited everyone in many
ways and has revolutionized the world. Privacy is an aspect which the researcher exhibits has a
very strong influence from the growing technological developments. Substantial privacy laws are
existent in the contemporary world which provides protection to individuals against unwanted
infringements into their lives and safeguards their rights for collection, storage or propagation of
personal information1. There is an emerging body of work that is searching at solutions to the
dilemma of keeping privateness while employing artificial intelligence and locating ways
wherein artificial intelligence can aid and fortify privacy. The artificial intelligence poses threat
to privacy which is very obvious but if this defensive outlook is changed then this element of
threat might become a solution to the privacy issue and thus there is an ardent need of
reprogramming privacy laws to utilize the technology.2 The paper aims at establishing the impact
of artificial intelligence on the privacy laws around the world and suggests that how the future
abstract of the legislations should be for it to sync with the AI and subsequent technologies.

1
Fred H. Cate & Rachel Dockery, Artificial Intelligence and Data Protection: Observations on a Growing
Conflict,77 ABY, LJ-987 (2018)
2
M Deora, An AI-entrusted future must have a stringent privacy & data protection law Read more at:
//economictimes.indiatimes.com/articleshow/64499489.cms?
from=mdr&utm_source=contentofinterest&utm_medium=text&utm_campaign=cppst ( Jun. 7, 2018),
https://economictimes.indiatimes.com/news/economy/policy/view-an-ai-entrusted-future-must-have-a-
stringent-privacy-data-protection-law/articleshow/64499489.cms?from=mdr.
PRIVACY

Privacy is the right to be left alone or to be free from misuse or abuse of one’s personality. The
right of privacy is the right to be free from unwarranted publicity, to live a life of seclusion, and
to live without unwarranted interference by the public in matters with which the public is not
necessarily concerned. Privacy and privacy laws have become topics of debate nowadays;
private data is very sensitive and delicate nowadays with the emergence of web portals and
constantly developing technology. Over 2.5 quintillion bytes of databases are collected every
day. Much of this information includes records that might permit people to be personally and
categorically recognized (or, non-public facts).Currently there are more than 2 million active
facebook users. Every minute approximately 5 lakh people share images via snapchat whereas
about 50,000 new images are added on the instagram web. There are 1/2 1,000,000 tweets sent
each minute. The quantity of private information this is being exchanged each day is
overwhelmingly growing. At the identical time that we proportion our private information,
there's growing subject with how that information is being gathered, stored, used and shared at
scale.3 Changes to the legal guidelines are being fuelled in part by the increasing public issue
with the idea of unfettered data collection and processing. The artificial intelligence systems play
an important role in privacy of data nowadays. The privacy and laws regarding it are discussed in
the paper in consonance with the AI technology.

ARTIFICIAL INTELLIGENCE

The term “artificial intelligence” (AI) describes the large aim of enhancing and featuring
computer systems to carry out responsibilities that typically require human knowledge and
interference, which includes visual notions and devices, speech recognition, decision-making,
and translation among languages4. This one term includes a wide style of technical
improvements, every of which may additionally provide demanding situations to the prevailing
data protection regulations. Artificial Intelligence systems have played a key role in many sectors

3
Y de Montjoye, , Solving Artificial Intelligence’s Privacy Problem, Special Issue 17 | 2017 Field
Actions Science Reports (Ali Farzanehfar ed., 2017).
4
A B, Rethinking Privacy for the AI era ( Mar. 27, 2019), https://www.forbes.com/sites/insights-
intelai/2019/03/27/rethinking-privacy-for-the-ai-era/#6d30ec3f7f0a.
be it health sector, transportation, financial services, marketing, agriculture, education and
training, cyber security, public authority and public services and data protection. Since the
research primarily focuses upon the impact of artificial intelligence in the field of privacy and
subsequent data protection. It presents challenges as well as benefits. The benefits as well as
challenges of the Artificial Intelligence systems to the privacy and privacy laws are discussed
below-

ARTIFICAL INTELLIGENCE AS A THREAT TO PRIVACY

What makes AI crucial is its data amassing pace and automation. The pace at which AI does
computation is quicker than human beings and it could be improved via adding additional
hardwares. AI essentially is designed to utilize large information units for analysis and is the
most effective way to interpret reasonable quantity of records in much less time. Artificial
intelligence machine learning can take care of an assignment without supervision which
facilitates to improve evaluation performance. Undoubtedly these characteristics are
commendable however there may be a downside to this. All those features affect privacy in
many ways be it data manipulation, data outlining, identification and tracking of personal data
etc. Reflecting through various countries and places, one thread emerges: improvements in
technology and science have in large part been perceived as threats to privacy and often led
policymakers to search for, and clients to demand, additional privacy safeguards within the
regulatory arenas. This ideology on technology as an undertaking to existing notions of, and
safeguards for, data privacy is also reflective of the mindset of modern regulations and
policymaking5. Whether thinking about the consequences of Big Data technologies, sensor
networks and the Internet of Things, face-detecting technology, usually-on wearable
technologies with voice and video interfaces, virtual and augmented truth, or artificial
intelligence, many policy reports and regulatory analyses have found this issue to be the most
appalling. Through platforms like social media applications like whatsapp, instagram etc there
has always been a constant threat of data being stolen and missued by the third party. WhatsApp
after being obtained through Facebook modified its privateness coverage, and the users had been
notified that "WhatsApp" account information of users might be shared with "Facebook" to

5
Greenberg, A., “An AI That Reads Privacy Policies So That You Don't Have To,” Wired (9 Feb. 2018),
https://www.wired.com/story/polisis-ai-reads-privacy-policies-so-you-donthave-to/.
improve "Facebook" advertisements and merchandise studies and the users’ have been asked to
agree to the revised terms for persisted use of WhatsApp on or before September 25, 2016. In
view this improvement Karmanya Singh Sareen6 and some other filed a writ petition before the
Hon’ble High Court of Delhi contending that getting rid of the protection to privacy of data of
customers of "WhatsApp" and sharing the same with Facebook changed into in infringement of
fundamental rights of the users guaranteed beneath Article 21 of the Constitution. The Hon’ble
Delhi High Court while deciding upon the case ordered that if the users prefer to absolutely
delete the WhatsApp account, WhatsApp shall delete customers’ information absolutely from its
servers and refrain from sharing users’ information with Facebook, and to this point because the
users who favor to continue to be in "WhatsApp" are involved, the present facts/facts/details of
such users upto September 25, 2016 shall not be shared with "Facebook" or any one of its
organization corporations.

ARTIFICIAL INTELLIGENCE AS BOOM TO PRIVACY

Artificial intelligence (AI) has developed and evolved manifold during the recent times. Today,
AI equipments are being utilized by many individuals and group of individuals around the world.
The benefits of AI definitely cast a bright shadow to the growth and prosperity of many sectors.
AI has tremendous potential to be the solution to the privacy infringement issue the world is
encountering nowadays. In out this essentially protective stance that privacy law has taken
historically in regards to AI and subsequent technology, it is really very crucial to point that law
within the broader context of AI and other technologies regularly transcends its traditional
position as a constraint on behavior acting via the imposition of sanctions 7. In fields such as
intellectual property and antitrust, regulations seeks to interact with AI in an extra nuanced way
through allowing or in a few cases leveling preferred innovative or disruptive interest. With this
expertise of legal system as a functionally differentiated response system, and an
acknowledgment that legal responses to AI developments have to no longer be understood as a
simple stimulus-reaction mechanism, it is feasible to discover a series of response patterns whilst
analyzing the evolution of privacy laws vis-à-vis technological changes8. Conventional legal
6
Karmanya Singh Sareen v UOI, 2016 S.C.C. 5334 (India).
7
U Gesser, , recoding privacy law: reflections on the future relationship among law, technology, and
privacy, 130 Harvard Law Review (Ryan Budish ed., 2016).
8
J Oltsik, Artificial intelligence and cybersecurity: The real deal ( Jun. 25, 2018),
https://www.csoonline.com/article/3250850/security/artificial-intelligence-andcybersecurity-the-real-
methods for shielding private data while handling information, taking data management
decisions, and drafting agreements of data- sharing9, among different activities, are not time
economical and now feasible for the big database management. Artificial intelligence tools are
designed in compliance with the legal system prevailing, on the way to aid automate
information-sharing choices and thus protect large amounts of data in a more effective and
efficient way. The issue of data being stolen and the subsequent issues in handling, managing
and collection of data can be swiftly be dealt with using the AI technology. Adding AI softwares
for the privacy protection can be time and cost economical. It has been proved through various
studies around the world that how the usage of AI in data protection is reaping goods, many
techniques like specialized tools for information storage and accessibility management, as well
as superior tools for data interpretation and analysis, can work in tandem with legal, institutional,
and other processes attainment of data privacy. The AI enables encryption of data which helps in
storage of information, analysis of information and thus facilitates smooth management over the
data without any manual labor. Differential privacy is a unique process of statistically analyzing
the data with mathematical inbuilt tools.

FUTURE PROSPECTS OF PRIVACY

The future prospects of privacy go hand in hand with the emerging tools like AI. Many data
regulators, organizations, lawyers, and lecturers are running hard to locate ways to deal with the
hurdles presented by AI to data protection laws. These are important initiatives and very much
intrinsic in mild of the immediate want for users of databases to comply with current information
safety laws. The issue is so much appalling and relevant that any attempt to fix the issue is prey
to the risk of database protection or intervening with the advantages of AI 10. The privacy laws
and AI must go hand in hand as both are inevitable in the contemporary time. The researcher
recommends that law must take firm stance in this respect to enact full-proof regulations to have
optimum utilization of AI and avoid mal-practices that may happen through its means. Many
general recommendations for data protection can be made like there should be complete

deal.html.
9
R Reiland, Artificial Intelligence Is Now Used to Predict Crime. But Is It Biased? ( May. 8, 2018),
https://www.smithsonianmag.com/innovation/artificial-intelligenceis-now-used-predict-crime-is-it-
biased-180968337/.
10
P Stone, Artificial Intelligence and Life in 2030 ( Aug. 2, 2016),
https://ai100.stanford.edu/sites/default/files/ai100report10032016fnl_singles.pdf.
transparency of AI systems, the users should be allowed to opt out of the systems, the usage and
purpose of AI must be pre-decided and thus the scope for its scope should be limited, data should
only be collected out of the will of the owner. The Laws should utilize AI to create robust
responsive mechanisms, AI softwares must be recognized as possible evidences in legal matters
where E-Courts can also be developed using the AI mechanism 11. The data protection regulations
are dynamic and should remain the same, the laws must focus on the following issues-

 Establishing framework for harm- The losses and harms associated with the usage of AI
systems should be identified and the regulations should cover these problems 12. There
should be fair compensation and relief mechanism for the aggrieved and speed disposal
of these matters must take place to prevent greater harm due to the delay.

 Transparency and Redress- Wherever the artificial intelligence technology is used there
must an effective system of transparency in pace where the control over the data is
limited and sufficient encryption methods should be employed for it13. Data processing by
many individuals might cause harm to them and their data so having redressal facility
would be mighty helpful in reducing the risks associated with data handling by AI.

 Well-equipped system of risk management- Risk control is the manner of strategically

finding out harms and advantages that could happen from an activity. Risk control does
no longer modify rights or duties, however by means of assessing both the likelihood and
severity of harms and benefits, it enables agencies perceive mitigation strategies and
allows the most fulfilling final results that maximizes potential benefits at the same time
as reducing the chance of harms to that it falls within desirable limits. Data protection has
long dependent upon risk management as a method for adhering with legal necessities
and making sure that databases are processed correctly and that the fundamental rights
and pursuits of people are protected and duly implemented. Yet these risk control
strategies, whether undertaken by using organizations or professional regulators, haven't

11
J Salmon, A guide to blockchain and data protection (Hogan Lovells 2018).
12
Fred H. Cate & Rachel Dockery, Artificial Intelligence and Data Protection: Observations on a
Growing Conflict,77 ABY, LJ-987 (2018)
13
M Shah, , Data Protection & Privacy Issues in India (Economic Laws Practice 2017).
 been formal and structured and did not take gain of a number of the broadly usual
standards and tools of danger management in other regions.

 Data Stewardship System- Many data protection legal guidelines place a few or all the
obligation for safeguarding privacy on person records subjects via the use of consent and
word. Effective management of the large records sets utilized by most AI programs calls
for shifting more duty away from individuals and toward data users and records
collectors14, who need to be held responsible for how the process and use the data. The
popularity that processors have to be liable for reasonably foreseeable harms will create
an extensive incentive for extra care in their management and utilization of databases
through Artificial intelligence. It will also restrict the exercise of permitting processors to
evade accountability by providing prior notice and inferring consent. This will help in
reducing lessen the load leveled on individuals and instead direct their focus on data
processing job only when it necessary. Processors will be benefited as less number of
resources would be wasted on, or having the enhancement or application of AI limited by
way of, efforts to adhere with insufficient measures, along with notices that nobody reads
or adhering to terms of consent which can be frequently false.

 Laws must focus on Data Uses and Impacts- Often there exists a prudent decision for the
databases to be publicized, accrued, or created15. For analyzing and accessing the risks
associated through those databases requires identification of the purpose for which the
concerned databases might be used. Data utilized for one purpose, or motive or subject to
a specific set of regulations and protections may be both useful and suited 16, whereas
similar databases utilized in any other context or for some other purpose or without
sufficient protections can be both useless and undesirable. Thus the data protection
norms, as in the words of the U.S. President’s Council of Advisors on Science and
Technology, “should pay more attention towards the use and purpose of collection of data

14
Cate, F., P. Cullen & V. Mayer-Schönberger, Data Protection for the 21st Century World (Microsoft
2014).
15
Cate, F. & V. Mayer-Schönberger, Data Use and Impact Global Workshop (Center for Applied
Cybersecurity Research 2013).
16
Dwork, C., “Differential Privacy: A Cryptographic Approach to Private Data Analysis," Privacy, Big
Data, and the Public Good: Frameworks for Engagement at 296 (2014).
 rather than its assessment and analysis.”Paying heed to the uses and purposes of data
doesn't dilute the responsibilities and accountability regarding data collection neither
should a focus on consent in rare or sensitive occasions be deserted. Rather, in many
circumstances, a more specific, as well as prudent handling of important data flows and
more efficient privacy protection is usually attained with greater focus on responsible and
accountable use17. Through this more of a purpose and use based method, data handlers
might check the fittingness of a meant use of private data now not via focusing mostly on
the conditions under which the data was primarily collected, however rather at the in all
possible impacts of a proposed use and the dangers they level on individuals. Thus a
comprehensive focus on the use of databases is more appalling due to the fact most
individuals and institutions already reflect on consideration on uses whilst comparing
their comfortness with proposed data handling activities.

CONCLUSION
Privacy is a very important issue in the 21st century due attention and importance is being
paid to it through several enactments ,laws , guidelines, and cases but still there exits big
loopholes in the system which has hollowed the privateness of private data. It can be
stated that Artificial intelligence is the gift to the human beings by the human beings and
all the debate and concerns regarding its usage and application are illusionary as in order
to protect the databases and private information from AI mal-use, AI is required to be
employed. It is the need of the hour18. The proliferation of Artificial intelligence systems
are already yielding sizeable outcomes, however brings forth some important issues.
Efforts to manage and handle these issues in consonance with the prevailing laws and
guidelines increasingly show the bounds of these laws and their incapacity both for
shielding privacy and for aiding development and enhancement in an ever growing data-
structured financial system. As new AI systems are being innovated and developed, there
is an evident opportunity at hand and an increasing and unavoidable requirement to
analyze and observe the current data protection regulations and frameworks and thereby
effectively bring forth new laws and guidelines in consonance with the twenty-first
17
H Nissenbaum, , Privacy in Context (Stanford University Press 2009).
18
G Spencer, Revised Guidelines on the Protection of Privacy and Transborder Flows of Personal Data
( Feb. 9, 2013), http://oecd.org/sti/ieconomy/oecd_privacy_framework.pdf..
century needs. The privacy laws thus need to take a more non-defensive stance towards
induction of AI into the data protection norms. The future prospects of Privacy laws and
privacy in general are bright given the policymakers become more assertive towards
adoption of laws in respect to requirements of the contemporary time.