Open Access Research Article
INVESTIGATING THE RISE VARIED APPLICATIONS AND WORLDWIDE EXPANSION OF AI, WITH EMPHASIS ON CHATGPT WITH SPECIAL MENTION TO EUS AI REGULATION
Published Paper
Article Details
INVESTIGATING THE RISE, VARIED
APPLICATIONS, AND WORLDWIDE EXPANSION OF AI, WITH EMPHASIS ON CHATGPT: WITH
SPECIAL MENTION TO EU'S AI REGULATION
Authored By - Seathal
Devi
"The
real question is, when will we draft an artificial intelligence bill of rights?
What will that consist of? And who will get to decide that?"
- Gray Scott
Abstract
Artificial
intelligence (hereinafter “AI”) is one of the vital human inventions that are used by many
of us in our day-to-day lives, either professionally or personally. A wide
range of areas, like healthcare, energy consumption, climate change, financial
risk management, etc., were using AI to make their work easier. For example,
AI, including Machine Learning (hereinafter “ML”), Deep Learning (hereinafter “DL”), Natural Language Processing (hereinafter “NLP”), and Robotic Process Automation (hereinafter “RPA”), aids healthcare tasks: analyzing diverse data, reports,
medical images like X-rays, scans, ECGs, MRIs, etc., from multiple
organizations. Many of the AI supporters trust that it does the equivalent work
of an activated human brain, monastically. However, few of its vibrant
promoter’s state that AI can perform many tasks better than humans. Hence, it
needs to be regulated to ensure that it is used in a legal, ethical, moral, and
appropriate way. In this research, the evolution, usage, advantages, latest
developments like Chat-GPT, and implications of AI are listed out, along with
the governance of AI by the European Union (hereinafter “EU”), which includes a detailed study
of the AI Act, which is to be the first ever legislation for the enforcement of
AI. In this research, the true nature of AI is going to be
analysed, and a legal conclusion is going to be given about whether AI has to
be regulated or not.
Introduction
The law is
a gadget that regulates everything in an orderly manner. The law has the power
to ensure the safety and security of every act that is happening around us.
Without proper study and regulation of societal event, it is impossible to
detect anomalies and inconsistencies in society. The same applies to the people
using AI. This unexhaustive AI has been evolving since 1940. It has an enormous
capability corresponding to that of a human brain, which could analyse large
amounts of data, much faster and more accurate than humans. AI can also perform
repetitive tasks without getting exhausted or making errors. New tools of AI
have been increasingly invented day by day. According to a survey by Klynveld Peat Marwick Goerdeler (hereinafter “KPMG”), the adoption of AI is increasing across
every sector, including industrial, technology, and retail[1].
In many countries, AI tools reach many citizens with easy access. So, it is
highly prone to risks such as misusing,
hacking, committing any crime, violations of fundamental rights, threats to
data protection, etc. Hence, there is an urgent need for regulation of these AI
tools and the launching stakeholders or companies to ensure that data is
protected, and AI is not used for any unethical use, or will not decrease any job
opportunities or intellectual capabilities of humans. Even though AI has been developing since 1940,
in many developed countries, they are still at the beginning stage of its
regulation.[2] Many developing countries does not
even start to regulate AI, yet AI has been evolving dynamically. This research aims to analyze the
inherent characteristics of AI and provide a legal verdict on whether
regulation of AI is necessary. This research also takes a comprehensive approach by
examining EU’s AI regulation with the first ever AI Act.
Objectives
Ø To figure
out the real nature and scope of AI.
Ø To analyse
the economic, social, and legal advantages and disadvantages of AI tools.
Ø To examine
the expeditious evolution of AI, in various field including Chat-GPT.
Ø To study
how developed countries like the EU regulate AI.
Ø To examine
whether all countries, especially those with a high market for AI, need proper
regulation.
Preface to AI
AI is an intellectual knowledge that
is capable of studying, analysing, planning, creating, and storing various
information via an artificially man-made device, and the purpose of its
invention is to assist humans in various ways, either mechanically or
intellectually. It is also capable of taking decisions in any specific
situation and can also synthesise various pieces of information in an orderly
manner from various sources. Russell & Norvig rightly expressed as
“defining the AI is not easy; in fact, there is no generally accepted
definition of the concept”.[3]
The Merriam-Webster Dictionary
defines intelligence as the ability to learn, understand, or deal with new or
trying situations.[4] It
additionally characterizes AI as a field within computer science that focuses
on simulating intelligent behavior in computers.[5]
Many of us will have the opinion that AI is a new form of technological
development, but it is a very longstanding idea that actually originated in the
1940s as a way to describe human intelligence in a machine.[6]
However, the formal discussion was
made at a conference at Dartmouth College in Hanover, New Hampshire, in 1956,
where the term "" was coined by John McCarthy, one of the
"founding fathers" of AI, together with Alan Turing, Marvin Minsky,
Allen Newell, and Herbert A. Simon.[7]
How AI works: Inspired by the human brain
The core idea behind the working of
AI is mirroring human intelligence, as AI’s ultimate aim is to mimic humans and
exert an effective intelligence beyond human’s natural intelligence without
getting exhausting. AI functions akin to a human child's brain, learning
without comprehensive interpretation. Similar to how a child absorbs messages,
AI stores and repeats, accumulating experience to shape future thinking. The
same applies to machines that are programmed with AI. It has two essentials: an
internal way of representing the information and transforming that information
into the anticipated output.[8]
These two activities happen in a repetitive way, and as a result, an AI device
would gain experience and become capable of exhibiting its own intelligence or making
relevant decisions in a given situation. Machines could do this better than
humans, as they are capable of storing larger amounts of data and are also
capable of speedy recollection of particular data when required. For this
process, AI adopts ML[9], by
which they gather the input, handle it, and understand it.[10]
An advanced form of ML is deep learning via artificial neural networks, which
is used to deduct large numbers of samples or big data.[11]
AI devices also use algorithms and
the Internet of Things (hereinafter “IOT”) to exhibit the functions of
smart devices and share information among other AI devices.[12]
Beyond these AI applications, functions like search algorithms and computer
vision analyze virtual data. NLP aids human interaction via automatic speech
recognition. Autonomous agents engage with users, performing tasks for humans
or other programs.[13]
Applications
of AI in Various Fields
AI is used in various fields, as it
is not restricted only to computer applications. The uses are as follows: -
1. Transportation
In the case of transportation, the
future is predicted to be completely dependent upon driverless vehicles with a
high-tech computer program algorithm. The dependence on gasoline will also be
reduced. According to Statista, autonomous vehicles are expected to
grow between 2019 and 2030.[14]
By 2030, it is estimated that approximately 58 million units of these vehicles
will be sold globally.[15]
The top five countries in the world, such as the US, Japan, France, the UK, and
Germany[16],
support the launch of AI-installed automated vehicles.
2.
Robots
AI robots are a combination of
software, a mathematical algorithm, and a mechanical part that work
concurrently. They are examples of cyberphysical systems. Robots not only
include humanoid robots but also self-driving cars, drones, and other smart
devices. They work much more accurately and quickly than humans without getting
drained. There are various differences between using a normal machine and an AI
machine for work. A normal machine cannot deduce anomalies, whereas the AI can
deduce them and can resolve them even in the absence of human monitoring. These
are more useful for industrial purposes than household purposes. However,
recent advances in AI are leading to the emergence of a new class of robot. In
the next five years, our households and workplaces will become dependent on
robots.[17]
3.
Healthcare
Through ML, DL, analysis of big data,
and Computer Vision, AI plays a major role in the healthcare sector. Health
organisations have accumulated big data sets in the form of health records and
images, population data, claims data, and clinical trial data, which helps in
assisting, diagnosing, and monitoring the patient.[18]
For example, during the COVID pandemic, AI chatbots were used to find the
right for treating COVID patients.[19]
Many countries, like the USA, China, the UK, Israel, India, Finland, and
Germany, are supporting the development of AI in healthcare via funding for
AI-research.[20]
4.
Public
safety
An AI algorithm is used in
surveillance cameras for facial recognition, to identify traffic violations,
allow for more resourceful allocation of policing through crime predictions,
and to deduct any anomalies in human behavior on public roads and streets.
Smart programs like biometrics, smart cameras, and video surveillance systems
are installed on the streets to ensure public safety.[21]
5.
Legal
Field
Since 1956, AI has been used in the
field of law, where logic was introduced as a tool for drafting and
interpreting legal documents.[22]
AI is being used in the legal field to draft contracts, guess legal outcomes,
acclaim judicial decisions about sentencing or bail, perform due diligence to
uncover background information, generate results that forecast litigation
outcome, and use data points from past case law, win/loss rates, and a judge’s
history to be used for trends and patterns.[23]
Apart from this, with regard to dispute resolution, Quantitative or
algorithmic legal prediction ( hereinafter “QLP”) is used, which uses
the techniques of ML and NLP to make a legal analysis of a particular case.[24]
QLP can help lawyers and judges make more informed decisions by providing them
with data-driven insights into the likely outcomes of legal cases. There are
many QLP applications, which include Correctional Offender Management Profiling
for Alternative sanctions. (hereinafter “COMPAS”)[25]
is a decision support tool used by many States in the US that assesses the
likelihood of recidivism in a criminal defendant. The Harm Assessment Risk Tool
(hereinafter “HART”) performs a similar function as that of COMPAS, yet
it employs a method called random forests, a form of ML used by the Durham
Constabulary in the United Kingdom.[26]
These tool lists are not exhaustive. Apart from these tools, there are
many others that are used to obtain quick legal advice, improve the efficacy of
many daily tasks of advocates, find proper issues in each case of facts, or
predict the cost of damages or compensation in a particular case. Yet, there
are many disadvantages to the use of these AI tools. High risk is involved if
these tools are fed biased or wrongful information, which in turn gives biased
results. They are also prone to hacking, and it can be difficult to interpret
the results of QLP models because they are often based on complex algorithms
that are difficult to understand.[27]
Recent Developments in AI: Chat-GPT
AI's evolution began in 1940. The
inaugural program, Logic Theorist (1955), by Simon and Newell, proved
mathematical theorems effectively.[28]
AI was also applied to some games, such as Nim, checkers, and chess, in the
early years of its development, around the 1970s.[29]
Nevertheless, following 1970, symbolic algorithms and expert systems,
alternatively referred to as knowledge systems, were created with the ability
to encode and manipulate knowledge specific to particular domains.[30]
From 1990, practical adoption of machine learning and deep learning enables
machines to enhance performance through data-based learning.[31] At present, we are at the advanced stage of
developing generative AI. Generative AI is a type of AI structure that is
capable of generating something in the system programmed with Large Language
Models (LLM), for example, text, code, image, video, etc. Even though
generative AI tools have been launching since the 1960s, like ELIZA[32] ,
Teaser AI[33] ,
Studio Bot[34]
, etc. At present, Open AI, a non-profit research company, has created
Chat-GPT, a generative AI tool that is expected to revolutionise the economy of
the world as it has many special features, in the Open AI research lab based in
San Francisco, CA.[35]
The term Chat-GPT has a meaning that is syllabified into Chat (which means
chatbot) and GPT (which means Generative pretrained transformer). Chat-GPT
excels in human-like responses to natural language. It answers questions,
generates text, translates languages, aids grammar, provides info, assists in
creative tasks, automates repetitive chores, and engages in conversation.
Chat-GPT is subject to a content policy that states that it cannot provide
assistance with illegal activities, including harm, or any form of criminal
behavior. It cannot provide assistance or advice that may pose a risk to an
individual's well-being, including self-harm, suicide, or harm to others. It
aims to use language that is respectful, inclusive, and free from
discrimination or offensive content, and cannot generate or promote hate
speech, discrimination, or any form of harassment. It can create graphics,
music, programmes, and images if fed with proper input.[36]
It mainly helps entrepreneurs and investors.[37]
Chat-GPT streamlines customer service and sales, cutting costs. Students and
graduates use it for presentations and job readiness, enhancing employability.
It analyzes stocks, but overuse may impact human creativity, values, and
intellectual recognition.[38]
It also has many limitations, as it
sometimes creates plausible-sounding but incorrect or nonsensical answers. It
also has a tendency to provide inaccurate information confidently as it is
updated to September 2021. For example, if you asked chat-GPT when it got
launched, it would probably give a response, as I, as an AI language model
known as GPT-3.5, was launched by OpenAI in June 2020. But to the real
fact, the Chat-GPT was launched by Open AI on November 30, 2022.[39]
In addition to this, it has the risk of privacy issues and may also be prone to
hacking. There is also a high expectation in society that these generative AI
tools like chat-GPT would decrease job opportunities, as they have various
abilities to give human-like responses to any natural language input.
Despite the above criticisms,
chat-GPT, being a tool that could quickly generate high-quality contents, is
being used by many in the world. As per statistics, chat-GPT is being used by
over 100 million users.[40]
It has created a record as it reached 1 million users within 5 days, which has
never happened so far in history. Other famous applications such as Instagram,
Facebook, and Netflix reached 1 million users in 2.5 months, 10 months, and 3.5
years, respectively.[41]
It is predicted that the revenue of Open AI will be $200 million by the end of
2023, and it will reach up to $1 billion by the end of next year.[42]
Chat-GPT is available in 56 countries.[43]
Out of that, 37 were developed countries, 16 were developing countries with
high populations, including India and China, and it has not even launched in
one under-developed country. Also, Open AI, being a non-profit research
company, has targeted countries with a high population in order to obtain a
high number of users.
Recent
developments in Chat-GPT: Chat-GPT-4
The chat-GPT was launched on 22,
November 2022 by the open AI. Within four months it has undergone enormous
changes and updates. Now we have the advanced version of chat-GPT i.e.,
Chat-GPT 4. The main difference between chat-GPT and chat-GPT 4 is that,
earlier version of Chat-GPT accepts only text as its input, whereas the latest
version accepts images, text, and also have the capability of returning
textual responses.[44]
This latest version was launched on 14th March 2023 by the Open AI.[45]
Within the span of this four months, many versions of chat-GPT, like 3, 3.5
were also launched.[46]
The chat-GPT 4 can do various tasks that the earlier versions failed to do,
including chat-GPT 3.5. chat-GPT 3.5 is a text-to-text model, whereas chat-GPT
4 is a data to text model, which means that it can generate text by analysing
the data.[47] In
addition to this chat-GPT 4 can also translate up to 26 languages for the valid
input.[48]
Distinguishing versions relies on parameters, machine learning model variables
for data adjustment. Neural networks' weights and biases are parameters.
Previous ChatGPT updated to September 2021, while ChatGPT-4 incorporates
current data. Bing AI, also known as Bing Chat, is an AI chatbot developed by
Microsoft and released in 2023.[49]
Bing AI, based on OpenAI's GPT-4, functions as a chat tool akin to ChatGPT 4.
Its Image Creator is multilingual.
Does AI
needs regulation?
AI's swift advancement, like OpenAI's
ChatGPT, brings potential and peril. AI tools enhance efficiency yet may
displace jobs. However, they could also generate fresh roles, facilitating
cost-effective product innovation. Balancing these impacts becomes crucial in
harnessing AI's potential.[50]
So, as a result, the implication of AI tools is uncertain in reality. Only
regulation will bring a clarity to it and these regulations should be built in
such a way, making AI tools to be benefited in a positive way. Also on the
other side, they are vital for the economic growth of the country.[51]
AI's boundaryless, algorithmic nature
and non-personhood pose regulatory challenges. Government oversight of
stakeholders and companies becomes crucial to address anomalies. Another risk
is that even the manufactures, are not aware of the result of the AI
technology.[52] If this
above fact was taken into consideration, then even the stakeholders and
companies launching AI cannot be blamed if it was misused or went
uncontrollable. There is also a risk of probable human rights violation and as
a responsible government, it has to ensure the rights of its citizens.
Sophisticated AI systems often
produce authentic content but may err by generating incorrect or copyrighted
material, violating intellectual property rights. Thus, advocating government
regulation for AI becomes reasonable.
The EU's attempt of regulating AI- The AI Act
The EU Commission proposed the AI Act
on April 21, 2021.[53]
The Act has attempted to define AI, which has much clarity in it.[54]
This act has adopted the risk-based regulation of AI, which means that the Act
classifies AI tools or applications based on the nature of the risk they can
cause to society. The AI applications are classified as either unacceptable
risk, high risk, limited risk, minimal risk, or no risk.
Unacceptable risk tools, labeled as
banned applications, pose human dangers. These encompass voice-activated toys,
discriminatory social scoring devices, and real-time biometric systems, such as
facial recognition.[55]
The
applications with high risk are as follows:
- Biometric identification refers
to the techniques that are used to identify human fingerprints: facial
recognition, voice recognition, vein scanning, and iris scanning.[56]
These are also used for the categorization of natural persons. These can
be dangerous when coupled with AI, as they store valuable human data of
humans.[57]
- Management and operation of
critical infrastructure to ensure the security of the Nation and the
public.[58]
- Education and occupational
training for the AI.
- AI in Health Care.
- Enforcement of Legal Actions and
the justice delivery system.
- Migration and border control
management This involves identity management, Border Management
Information Systems (BMIS), Integrated Border Management (IBM), and Humanitarian
Border Management (HBM).[59]
- Assistance in the legal
interpretation and application of the law.
The above
list is not exhaustive. These high-risk AI applications will undergo strict
reviews under the Act. For example, they would undergo Conformity assessments,[60] in which the
following are analysed:
ü Data that is fed
ü Tools, and its nature.
ü The presumptions and biases
ü Interacts
ü Outputs
ü Overall performance
Results are
rigorously evaluated for clarity, adequacy, and transparency before market
release, with subsequent post-market monitoring. Low-risk AI tools maintain
transparency standards, aiding user assessment of relevance. This Act has also
been amended in a way that can regulate generative AI like ChatGPT.[61] The
main aim of this act is to ensure transparency f and prevent it from generating
illegal content and publishing copyrighted data.[62]
The AI Act, which is of global
standard, is also drafted for the constitution of the European AI Board.[63]
The primary objective of this Act is to ensure the safety, transparency,
traceability, non-discrimination, and environmental friendliness of AI systems
employed within the EU.[64]
The Member of European Parliament (MEP) negotiated the AI bill on June 14, 2023[65]
, where they announced an authoritative ban on AI, which is involved in
the tasks of biometric surveillance, emotion recognition, and predictive
policing.[66] The
generative AI systems like ChatGPT were also deliberately regulated so that
they should be updated in such a way to disclose the contents that they were
generated by AI, expressly.[67]
After the negotiations with the European Council, the AI Act is expected to be
enforced.[68]
Conclusion
The fact is, even most of the
developed countries were at the beginning of the regulation. Many countries are
not aware of the true nature of AI and its dangerous implications. Many judges,
lawyers, and judicial officers around the world are unable to make judgements
regarding liability in cases involving AI disputes. AI is neither identified as
a juristic person nor as a natural person. Its characters are
undistinguishable. For example, in the case of Gonzalez v. Google LLC[69],
the US Supreme Court unanimously held that the family’s claims against the
social media companies were not allowable under the Antiterrorism Act, and this
judgement was favourable to Google, even though Google has been used by
terrorist activities, and as a result, the petitioner lost his daughter. The
same happened in the case of Twitter, Inc. v. Taamneh.[70]
However, in the case of Gonzalez v. Google LLC, Google cannot be held liable as
even as an inventor or stakeholder as Google is unaware of the results of the
AI installed in its applications, and a person cannot be held liable unless he
has a Mens rea and Actus rea. However, mere immunity cannot be given to
companies and stakeholders without proper regulation.
Therefore, AI needs a regulation, not
only for the developed countries but also for the developing countries like
India and China, where the population is very high. As the population
increases, the risk increases along with it. This paper strongly suggests
national regulation of AI, as each country is completely unique in its social,
economic, political, legal, and cultural identities. National regulation can
also have a great impact on international regulation if it is binding. This
paper appreciates the efforts that were taken by the EU countries to regulate
AI, and it inspires many countries around the world.
[1]Embracing the rapid pace of AI, available
at: https://www.technologyreview.com/2021/05/19/1025016/embracing-the-rapid-pace-of-ai/ (Visited on June 20, 2023).
[3] Philosophy of artificial
intelligence, available at: https://en.wikipedia.org/wiki/Philosophy_of_artificial_intelligence, (Visited on June 22, 2023).
[4] Intelligence, available at:
https://www.merriam-webster.com/dictionary/intelligence
(Visited on June 22, 2023).
[5] AI, available at: https://www.merriam-webster.com/dictionary/artificial%20intelligence
(Visited on June 22, 2023).
[6] Tanya Lewis, “A Brief History of
AI”, LiveScience, Dec. 5, 2014.
[7] Dartmouth Conference, AI in 1956
and Reflections After 50 Years, available at: https://kikaben.com/dartmouth-conference-ai-in-1956/,
(Last Modified June 16, 2023).
[8] Jan De Bruyne and Cedric
Vanleenhove (eds.), Artificial Intelligence and The Law, 4, (Intersentia
Ltd, UK).
[9] Tom M. Mitchell, Machine
Learning, 14-16, (McGraw-Hill, USA, 1997).
[10] Jan De Bruyne and Cedric
Vanleenhove (eds.), Artificial Intelligence and The Law, 1, (Intersentia
Ltd, UK).
[11] Jan De Bruyne and Cedric
Vanleenhove (eds.), Artificial Intelligence and The Law, 7, (Intersentia
Ltd, UK).
[12] Rajiv Malhotra, Artificial
Intelligence and the future of the power, 43, (Rupa Publications India Pvt.
Ltd, New Delhi 110002, 1st edn., 2021).
[13] Jan De Bruyne and Cedric
Vanleenhove (eds.), Artificial Intelligence and The Law, 11-14,
(Intersentia Ltd, UK).
[14] Projected sales of autonomous
vehicles worldwide from 2019 to 2030, available at: https://www.statista.com/statistics/1230733/projected-sales-autonomous-vehicles, (Visited on June 24, 2023).
[15] Projected sales of autonomous
vehicles worldwide from 2019 to 2030, available at: https://www.statista.com/statistics/1230733/projected-sales-autonomous-vehicles, (Visited on June 24, 2023).
[16] The top five best-equipped
countries to support autonomous vehicles – Who’s leading the self-driving
revolution?, available at: https://electrek.co/2022/03/04/the-top-five-best-equipped-countries-to-support-autonomous-vehicles (Visited on June 24, 2023).
[17] These 5 robots could soon become
part of our everyday lives, available at: https://www.weforum.org/agenda/2022/02/robots-future-tech/
(Visited on June 24, 2023).
[18] What is artificial intelligence in
healthcare?, available at: https://www.ibm.com/topics/artificial-intelligence-healthcare, (Visited on June 25, 2023).
[19] A.S. Miner et al, ‘Chatbots in the
fight against the COVID-19 pandemic’, III, Digital Medicine, 1–4, (2020).
[20] Indiaai, Report: Artificial
Intelligence in Healthcare Sector, (November 2022).
[21] Christopher Rigano, “Using
Artificial Intelligence to Address Criminal Justice Needs”, Mckinsey,
Mar. 10, 2020.
[22] L.E. Allen, ‘Symbolic logic: A
razor-edged tool for drafting and interpreting legal documents’, LXVI, Yale
L.J., 833–879, (1957).
[23] Law Bots: How AI Is Reshaping the
Legal Profession, available at: https://businesslawtoday.org/2022/02/how-ai-is-reshaping-legal-profession/, (Visited on June 25, 2023).
[24] J. Hafner et al., “Quantitative
Legal Prediction - or - How I Learned to Stop Worrying and Start Preparing for
the Data-Driven Future of the Legal Services Industry,” XXVII, Artificial
Intelligence and Law, 337-359, (2019).
[25] COMPAS (software), available at:
https://en.wikipedia.org/wiki/COMPAS_%28software%29, (Visited on June 25, 2023).
[26] UK police are using AI to inform
custodial decisions – but it could be discriminating against the poor, available
at: https://www.wired.co.uk/article/police-ai-uk-durham-hart-checkpoint-algorithm-edit, (Visited on June 25, 2023).
[27] The advantages and disadvantages
of AI in law firms, available at: https://venturebeat.com/datadecisionmakers/the-advantages-and-disadvantages-of-ai-in-law-firms/, (Visited on June 25, 2023).
[28] History of Artificial Intelligence
– AI of the past, present and the future!, available at: https://data-flair.training/blogs/history-of-artificial-intelligence/, (Visited on June 26, 2023).
[29] “A Brief History of Artificial
Intelligence” Live Science, Dec. 5, 2014.
[30] Timeline of AI: a brief history of
artificial intelligence it’s highlights, available at: https://www.ai.nl/artificialintelligence/timeline-of-ai-a-brief-history-of-artificial-intelligence-its-highlights/, (Visited on June 26, 2023).
[31] Evolution of Artificial
Intelligence, available at: https://inlocrobotics.com/en/evolution-of-artificial-intelligence/, (Visited on June 26, 2023).
[32] Josh
Fruhlinger, What
is generative AI? The evolution of artificial intelligence, Infoworld,
(2023).
[33] Teaser AI: Can Generative AI Solve
the Ghosting Problem in Online Dating?, available at: https://www.thequint.com/tech-and-auto/tech-news/teaser-ai-dating-app-chatbot, (Visited on June 25, 2023).
[34] Generative AI, available at:
https://ai.google/discover/generativeai, (Visited on June 25, 2023).
[35] Matthew Kupfer, “I Tried to Visit
OpenAI’s Office. Hilarity Ensued”, The San Francisco Standard, Dec. 20,
2022.
[36] Thomas H. Davenport and Nitin
Mittal, How Generative AI Is Changing Creative Work, Harvard Business Review,
(2023).
[37] ChatGPT is about to revolutionize
the economy. We need to decide what that looks like, available at: https://www.technologyreview.com/2023/03/25/1070275/chatgpt-revolutionize-economy-decide-what-looks-like/, (Visited on June 25, 2023).
[38] ChatGPT is about to revolutionize
the economy. We need to decide what that looks like, available at: https://www.technologyreview.com/2023/03/25/1070275/chatgpt-revolutionize-economy-decide-what-looks-like/, (Visited on June 25, 2023).
[40] Number of ChatGPT Users (2023), available
at: https://explodingtopics.com/blog/chatgpt-users, (Visited on June 25, 2023).
[41] Number of ChatGPT Users (2023), available
at: https://explodingtopics.com/blog/chatgpt-users, (Visited on June 25, 2023).
[42] Jeffrey Dastin, Krystal Hu and
Paresh Dave, “Exclusive: ChatGPT owner OpenAI projects $1 billion in revenue by
2024”, Reuters, Dec. 15, 2022.
[43] Chat GPT Available Countries List,
available at: https://chatgptdetector.co/what-countries-is-chat-gpt-available/, (Last Modified March 29, 2023).
[44] GPT-4 vs. ChatGPT: An exploration
of training, performance, capabilities, and limitations, available at: https://towardsdatascience.com/gpt-4-vs-chatgpt, (Visited on June 25, 2023).
[46] WHO OWNS OPENAI CHATGPT AND WHEN
DID IT LAUNCH?, available at: https://www.hitc.com/en-gb/2022/12/07/who-owns-openai-chatgpt-and-when-did-it-launch/, (Visited on June 25, 2023).
[47] Eric Griffith, “GPT-4 vs.
ChatGPT-3.5: What’s the Difference?”, PC, Mar. 16, 2023.
[48] Eric Griffith, “GPT-4 vs.
ChatGPT-3.5: What’s the Difference?”, PC, Mar. 16, 2023.
[49] Microsoft Bing, available at:
https://en.wikipedia.org/wiki/Microsoft_Bing, (Visited on June 25, 2023).
[50] Marcin Fr?ckiewicz, “The Economics
of Chat GPT-4: Analyzing the Financial Implications of Advanced AI”, TS2,
Mar.16, 2023.
[51] Chia-Hui Lu,“The impact of
artificial intelligence on economic growth and welfare”, Journal of
Macroeconomics, LXIX 3 (2021).
[52] Diksha Munjal, “Explained | What
is the EU’s Artificial Intelligence Act and how does it plan to rein in tech
like ChatGPT?”, TH, May.3, 2023.
[53] Artificial intelligence act, available
at: https://www.europarl.europa.eu/, (Visited on July 1, 2023).
[54] Benjamin Mueller, “The Artificial
Intelligence Act: A Quick Explainer”, Datainnovation, May.4, 2021.
[56] AI-Powered Biometric
Authentication, available at https://www.iotforall.com/press-releases/ai-powered-biometric-authentication, (Visited on July 2, 2023).
[57] EU AI Act: first regulation on
artificial intelligence, available at https://www.europarl.europa.eu/news/en/headlines/society/20230601STO93804/eu-ai-act-first-regulation-on-artificial-intelligence, (Visited on July 2, 2023).
[58] critical infrastructure, available
at: https://www.techtarget.com/whatis/definition/critical-infrastructure, (Visited on July 2, 2023).
[59] Border Management, available at:
https://www.iom.int/sites/g/files/tmzbdl486/files, (Visited on July 2, 2023).
[60] Guidance on Classification and
Conformity Assessments for High-Risk AI Systems Under EU AI Act, available
at: https://ai-regulation.com/guidance-on-high-risk-ai-systems-under-eu-ai-act/, (Visited on July 2, 2023).
[61] EU AI Act: first regulation on
artificial intelligence, available at https://www.europarl.europa.eu/news/en/headlines/society/20230601STO93804/eu-ai-act-first-regulation-on-artificial-intelligence, (Visited on July 2, 2023).
[62] Christopher McFadden, “EU
officials confident its 'AI Act' will pass sometime this year”, Interesting
Engineering, May.1, 2023.
[63] James Vincent, “The EU is
considering a ban on AI for mass surveillance and social credit scores”, Theverge,
Apr.14, 2021.
[64] EU AI Act: first regulation on
artificial intelligence, available at https://www.europarl.europa.eu/news/en/headlines/society/20230601STO93804/eu-ai-act-first-regulation-on-artificial-intelligence, (Visited on July 2, 2023).
[65] MEPs seal the deal on Artificial
Intelligence Act, available at, https://www.euractiv.com/section/artificial-intelligence/news/meps-seal-the-deal-on-artificial-intelligence-act/,
(Last Modified May 23, 2023).
[66] MEPs ready to negotiate first-ever
rules for safe and transparent AI, available at https://www.europarl.europa.eu/news/en/press-room/, (Visited on July 2, 2023).
[67] MEPs ready to negotiate first-ever
rules for safe and transparent AI, available at https://www.europarl.europa.eu/news/en/press-room/, (Visited on July 2, 2023).
[68] EU Lawmakers Pass Landmark AI
Regulation Bill, available at, https://foreignpolicy.com/2023/06/14/eu-ai-act-european-union-chatgpt-regulations-transparency-privacy, (Visited on July 2, 2023).
[69] Lawrence Hurley, “Supreme Court
sidesteps ruling on scope of internet companies' immunity from lawsuits over
user content”, NBCnews, May.19, 2023.
[70] Adi Robertson, “The Supreme Court
will determine whether you can sue platforms for hosting terrorists”, Theverge,
Oct.3, 2022.
Article Information
INVESTIGATING THE RISE VARIED APPLICATIONS AND WORLDWIDE EXPANSION OF AI, WITH EMPHASIS ON CHATGPT WITH SPECIAL MENTION TO EUS AI REGULATION
Authors: Seathal Devi
- Journal IJLRA
- ISSN 2582-6433
- Published 2023/10/20
- Issue 7
About Journal
International Journal for Legal Research and Analysis
- Abbreviation IJLRA
- ISSN 2582-6433
- Access Open Access
- License CC 4.0
All research articles published in International Journal for Legal Research and Analysis are open access and available to read, download and share, subject to proper citation of the original work.
Disclaimer: The opinions expressed in this publication are those of the authors and do not necessarily reflect the views of International Journal for Legal Research and Analysis.