AI, MACHINE LEARNING & BIG DATA LAWS & REGULATIONS 2022 | INDIA
Computing devices are able to mimic human behaviour, to an extent, through ‘artificial intelligence’. ‘Artificial intelligence’ (AI) is the decision-making ability of a machine, which often involves the processing of large amounts of data, literally ‘big data’ (BD), by the use of algorithms. This ‘big data’ can be used to develop ‘intuitive learning’ or ‘thinking’ in a machine, i.e., ‘machine learning’ (ML). Evidently, the relationship between AI, ML and BD is inescapable.
India has a natural advantage in this field, coupled with an obvious requirement given the population – India has readily collectible large and diverse data, and also the technical ability to utilise such data. Fast-paced advancements in technology, excessive consumerism, and technological agility have contributed to the dynamic situation that we have today. In this chapter, the authors have analysed the current trends in India relating to ‘artificial intelligence’, ‘machine learning’ and ‘big data’, and have examined attendant legal aspects with respect to ownership, antitrust, data protection, governance and regulatory matters.
1.1. Artificial intelligence, machine learning and big data trends in India Why was involvement in AI made necessary?
AI is predicted to contribute $15.7 trillion to the global economy in 2030.1 We stand on the brink of a technological revolution led by AI which will fundamentally alter the way we live, work and relate to one another.2 AI has affected our lives more than we realise. From waking up to Siri’s news updates to falling asleep to a movie suggested by Netflix’s recommendation engine, the technology underlying the Fourth Industrial Revolution has penetrated our daily lives.3
Since other countries are making rapid progress in the field of AI, and with globalisation being inevitable, it is imperative that India begins to see AI as a critical element of national security strategy, focus on AI-based innovation and establish AI-ready infrastructure to prepare India’s jobs and skills market for an AI-based future to secure its strategic interests.4
AI, ML and BD trends
In the absence of any official BD repository and disclosure requirements regarding the manner of use of BD, it is difficult, at this juncture, to make an accurate assessment of any trend in India in this regard. However, analysing the flow of investments in the public and private sector, the following trends may be deduced:
I. Government thrust towards innovation and development of AI
The framework for regulating AI and its applications is in its embryonic stages and there is much to traverse. It is evident from the following that the Government is working towards creating an AI-friendly technological ecosystem in India:
a. In 2017, the Ministry of Commerce and Industry set up an AI taskforce which highlighted various sectors of importance5 for the AI regime and the challenges in adopting AI in India.
b. In 2018, the National Institution for Transforming India (NITI Aayog)6 was directed to initiate AI programs and their applications. The Ministry of Electronics and Information Technology (MeitY) constituted four committees to develop a policy framework and analyse issues like leveraging AI, key policy enablers required across various sectors, and legal and ethical issues regarding AI.7
c. In January 2020, NITI Aayog recommended that an AI-explicit computer framework ‘AIRAWAT’8 be set up for satisfying the processing needs of innovation hubs, start-ups,
AI researchers and students.9
d. In February 2021, NITI Aayog published a paper called ‘Responsible AI for All’ where it proposed the following principles10 for the responsible management of AI systems that may be leveraged by relevant stakeholders in India:
i. Principle of Safety and Reliability: AI should be deployed reliably as intended and sufficient safeguards should be adopted in order to ensure that the safety of stakeholders is maintained and their grievances are addressed. It also proposed the monitoring of AI-based systems on a regular basis to assure that it performs its functions in a manner, as it had intended.
ii. Principle of Equality: AI systems must treat individuals under the same circumstances relevant to the decision equally.
iii. Principle of Inclusivity and Non-discrimination: AI systems should not deny opportunity to a qualified person on the basis of their identity. In case of an adverse decision, an appropriate grievance redressal mechanism should be designed in a manner affordable and accessible to everyone irrespective of their background.
iv. Principle of Privacy and Security: The fundamental right of privacy and security of the data of individuals shall be ensured at all times and access to the AI systems shall only be provided to the authorised persons.
v. Principle of Transparency: The design and functioning of the AI system should be recorded and made available for external scrutiny and audit to the extent possible to ensure the deployment is fair, honest, impartial, and guarantees accountability.
vi. Principle of Accountability: All stakeholders involved in the design, development and deployment of the AI system must be responsible for their actions.
vii. Principle of Protection and Reinforcement of Positive Human Values: AI should promote positive human values and not disturb social harmony (in any way) in community relationships.
II. Pioneering efforts of the private entities in the AI sector
Private initiatives in India have been far ahead in the development and use of AI compared to the Government. From utilising various applications powered by AI to providing various online services like MakeMyTrip, Firstcry and Flipkart, which learn from consumer’s online behaviour for making intelligent goods and services suggestions, corporates have been engaging in the use of AI for a long time. Big conglomerates are infusing AI to automate day-to-day operations. The insistence on the automation of daily tasks is further necessitated by the fast growth of business.
Indulgence of the private entities in AI is evident from the investments being made by them, specifically in the areas of e-commerce, anomaly detection, banking and finance, and retail. Flipkart uses AI-powered robots at sortation centres to process 4,500 shipments an hour with twice the speed and 99.99% accuracy. Swiggy uses AI-powered chatbots for customer support and an AI-ML model for search result optimisation.11
In India, many large corporations like Google and Walmart Labs are acquiring small startups for their AI innovations. Investments by private entities in AI-specific start-ups and the facilitation of an AI-friendly ecosystem by Government initiatives has resulted in the blossoming of AI start-ups.
It is noteworthy that the developments pursued by the Government in AI are primarily in collaboration with private entities. For instance, NITI Aayog’s collaboration with IBM for developing precision agriculture using AI for doubling farmers’ income by 2020, by using a machine learning process along with different computer algorithms for crop classification and area estimation.12 Additionally, the Government of Andhra Pradesh collected information from a range of databases, and processed the information through Microsoft’s Machine Learning Platform to monitor children and devote student-focused attention on identifying and curbing school drop-outs.13
1.2. Applications of AI
The integration of AI in our lives is affected by several factors, including: the digital divide; inequitable internet access; local economy; geographical location; and not forgetting the ‘culture’ and ‘adaptability’ quotient.
Private and public sectors have perceived the abundant applications of AI, ML and BD and have begun their endeavours to exploit them. Permutations and combinations of different types of ML14 are used to suit the purpose of the program.
I. Healthcare sector of India
a. In 2018, India ranked 145th in terms of healthcare access and quality rating.15 India also has a high rate of health issues, particularly diabetes; 19% of patients16 in the world are of Indian nationality. One of the consequences of diabetes is diabetic retinopathy which, if left untreated, can lead to blindness. Microsoft and Forus Health are working with NITI Aayog on a device called 3Nethra for early detection of diabetic retinopathy using AI-based retinal imaging APIs, which uses cloud intelligence to more closely assess eyes. The system automatically grades the images and identifies if the patient has diabetic retinopathy.17 It is a blessing for a country like India, where there are only
20,000 ophthalmologists for 1.3 billion people.18
b. An Indian healthcare start-up called NIRAMAI has developed a device using ML to detect breast cancer in women at a much earlier stage than traditional methods. It checks thermal images against positive and negative reports using BD analytics, AI and ML.
c. NITI Aayog is also under a discussion regarding a project called Imaging Biobank for Cancer. The project will use AI-based radiomics to improve decision-support in cancer treatment at low cost using image-based features.19
II. E-Commerce sector
a. It is expected that the Indian e-commerce market will grow to $200 billion by 2026, which may be attributed to increasing Internet and smartphone penetration.20 Using AI and ML algorithms, the online shopping experience has been personalised for every customer. They predict buyer behaviour based on past searches and orders and recommend products that would be relevant to the customer.
b. AI and ML are used by e-commerce entities to stock warehouses in accordance with preferences in a geographical area. Machine learning algorithms are used to predict future demand for products and accordingly fill shelves.
III. Defence sector
a. Apart from the huge success of AI in the automation of activities in the commercial sector, AI is being developed and infused in the defence and national security sector of India. The need for such implementation is aggravated by the fusion of AI into defence activities by other nations across the globe.
b. The reported developments of India involved in AI in the defence sector include:
i. Development of more than 200 DAKSH Robots, Remotely Operated Vehicles (ROV) which are used to defuse explosives, by the Indian Army.
ii. Development of RoboSen, which is a mobile robot created with the purpose of patrolling, reconnaissance and surveillance, by the Indian Army. RoboSen is capable of autonomous navigation in rough terrain with the ability to circumvent obstacles and provide continuous video feedback.21
iii. Development of Network Traffic Analysis (NETRA) by the Centre for Artificial Intelligence Research (CAIR) to monitor traffic on the Internet. NETRA is capable of analysing voice traffic going through video conferencing applications and can intercept messages with specific keywords for reconnaissance and intelligence collection purposes.
1.3. How has COVID-19 affected these trends, if at all?
I. Accelerated the growth of AI
The pandemic that hit the world two years ago has acted as an accelerant for the adoption of AI across all sectors. Businesses have adopted AI-powered tools to overcome the limitations of minimal contact, necessitated on account of COVID-19. AI-enabled tools have enabled reduced manpower deployment in several areas.
AI-enabled tools have allowed businesses to address changing consumer requirements and thus consumer behaviour. AI-powered cognitive assistants have been effectively deployed to perform back-office tasks like ordering new credit cards, issuing refunds or cancelling orders.
These tools have enabled organisations to reopen their manufacturing and office locations in the post-pandemic world by allowing organisations to implement best practices from a health, hygiene and safety perspective and comply with the Standard Operating Procedure. The remote workforce of each organisation is also relying more on AI-enabled digital assistants to do their work without loss of productivity.
As past knowledge and experience may fail to provide the right insights in these times of disruption, organisations are increasingly making decisions based on current data using AIenabled predictive and prescriptive tools.
According to a global study by PwC India,22 the highest increase in the use of AI during COVID-19 has been witnessed in India with 94% of over 200 chief executives in India surveyed stated that they have either adopted or are planning to implement AI in their companies. This study deduced that AI adoption in India was seen to be led by the travel and hospitality sector with 89% of the surveyed firms doing so, followed by telecoms, media & technology firms (86%), financial services (82%), and healthcare and pharma companies (73%) adopting it during the year. II. Paved the way for development in various fields
Government, companies, and start-ups in India are resorting to innovative means to combat the spread of the virus, and also address the challenges posed by the virus. This includes: the adoption of AI-enabled CCTV surveillance systems to detect the lack of face mask wearing and social distancing norms; AI-powered devices to record persons’ temperatures, contactless, at a distance; and deployment of AI technologies to screen and monitor the disease in hospitals, etc.
The situation brought about by COVID-19 has brought to the forefront the need for advancement in several areas, such as: use of AI to introduce e-wills; or the need for AI in socially distanced education, learning and conducting of exams. The following are some of the developments by public and private entities brought about by the advent of the pandemic:
i. The Ministry of Electronics and Information Technology introduced an AI-enabled Corona Helpdesk called ‘Haptik’, which essentially is an AI-powered WhatsApp chatbot and helps individuals in resolving their problems.23
ii. The AI & Robotics Technology Park, set up by the Indian Institute of Science, Bangalore, along with the help of the Department of Science & Technology (DST), Govt. of India and HealthTech startup Niramai have developed an AI-driven platform called ‘XraySetu’.24 This platform helps patients in the early detection of COVID-19 but also provides them with an opportunity to share low resolution chest X-ray images over WhatsApp and consult the doctors for their treatment.25
iii. Researchers from IIT Jodhpur have also formed an AI-based solution for predicting COVID-19 only with the help of chest X-rays.26 The solution is enabled by a deep learning-based algorithm called ‘COMiT- Net’ which detects the chest X-rays that are affected by COVID-19 in comparison to those which are not.27 This algorithm predicts pneumonia along with the infected regions in the lungs.28
III. Raised awareness of issues like privacy, particularly in the context of public health
As one of the prescribed measures advised by the World Health Organization (WHO) to contain the spread of COVID-19, the Government decided to digitise contact tracing with the launch of the Aarogya Setu App. This app can be used as a mechanism to trace contacts, with the help of Bluetooth technology and AI, based on proximity with other devices that have the same app installed. Since the app seeks certain personal information from its users, concerns have been raised regarding privacy rights, more so in the absence of a data protection law in India. Recently, Karnataka High Court vide an Order dated 25th January 202129 restrained the Central Government from sharing the data collected through the App with other government departments and agencies, citing that informed consent of the users is limited to collection and manner of collection of information, use of information and retention, but no informed consent is taken with regard to the sharing of response data.
To sum up, COVID-19 has not only given an impetus to AI-enabled tools, but also brought attention to several relevant and urgent issues that must be addressed at a global level. COVID-19 has acted as a great leveller in the sense that it has brought into focus that mankind across the globe is susceptible regardless of geography, demography and status. The outbreak of COVID-19 has made us acknowledge the essence of interdependency worldwide and concede to the fact that personal well-being is intertwined with the ability of society as a whole to overcome the crisis, and AI strengthens this outlook by facilitating the use of vast and diverse data, irrespective of geography, demography and status to make informed decisions even in these uncertain times.
1.4. Key legal issues
‘With great power comes greater responsibility’ – a proverb which became popular during the French Revolution, essentially responsible for bringing an accountable and law-abiding form of Government, has never been more befitting than in the current AI revolution. ‘Accountability’ is essential to the continuous development of AI and its application. It is hardly possible to think about the applications of AI without the legal implications, the key elements of which are as follows: a. Ownership/protection:
i. Are AI applications to be categorised under copyright law and/or under patent law? ii. Is AI a mere ‘tool’ and therefore the owner of the ‘tool’ should be identified with the intellectual property generated by such ‘tool’, or the AI application itself is the creator of the intellectual property in question and therefore should be recognised as its owner?
iii. Should BD generically or at least in respect of certain areas be regarded as ‘critical infrastructure’, and if ‘yes’, is there a need to regulate access and use of the same? b. Antitrust/competition laws:
i. Whether or not access and use of BD has ascribed such an advantage to certain enterprises which causes are likely to cause an ‘appreciable adverse effect’ on competition in the market. ii. Whether or not non-price competitive factors should be a consideration for approval of proposed combinations.
iii. Whether or not collusion through AI applications are anti-competitive under competition law.
c. Board of directors/governance:
i. What is the impact of AI in the decision-making process at board level. ii. What is the role of AI in corporate governance.
2.1. With the rapid advancement in the field of AI and its applications that cover almost every aspect of our lives, the creators of AI have become aware of the need to have ownership of AI applications and to protect the rights attached with such AI applications. The AI applications may be categorised into and protected under the following categories:
2.1.1. The Copyright Act, 1957 (‘Copyright Act’) – Under Section 2(o) of the Copyright Act, both the ‘source code’ and the ‘object code’ of AI applications are protected as ‘literary works’. The author of the AI applications, i.e. the developer, is considered to be the owner of such AI application, except when the author generates or creates an AI application in the capacity of an employee during the course of employment. The Copyright Act also permits fair use and reverse engineering.
2.1.2. The Patents Act, 1970 (‘Patents Act’) – Under Section 3(k) of the Patents Act, computer programs are not patentable per se. An AI application may be patented if it is attached to an invention along with hardware and it is proven that hardware is an essential component of such invention along with the software. For instance, Google LLC filed patent application no. 3023/KOLP/2014 titled ‘Location History Filtering’ and was subsequently granted a patent after the examiner raised objections under Section 3(k) of the Patents Act. In response to the objection, the applicant proved that the claims were not related to computer programs but a computing device, which enhance its technical effect.
In July, 2021, a Standing Committee set up by the Parliament (‘Standing Committee on IPR Regime’) published a report titled ‘Review of Intellectual Property Rights Regime in India’ which proposed implementing a strong IPR framework in order to extract benefits from AI, and recommended an amendment to Section 3(k) of the Patent Act, 1970. In this regard, it was noted: ‘As per the provisions of this Section, a mathematical or a business model or a computer program or algorithms run by Artificial Intelligence are not patentable and that the precondition which requires the mandatory presence of a human inventor for innovating computer related inventions (innovations by AI and machine learning) hinders the patenting of AI induced innovations in India.’30
2.1.3. Trade secrets – In India, the design, idea and structure of an AI application may be protected as a ‘trade secret’ based on its nature and distribution through contracts or under the law of torts.
2.1.4. Licence agreements – Access to AI applications can be granted by way of licence agreements by the owners of such AI applications. Such licence agreements are broadly divided into two categories: exclusive licences; and non-exclusive licences. The Copyright Act recognises the concept of ‘exclusive licence’.31 Under the Copyright Act, an exclusive licensee enjoys the rights comprised in the copyright of a work which is akin to that of the owner and includes the right to prosecute, defend and enforce the intellectual property rights.
2.2. Ownership of intellectual property created by AI applications – Under the Copyright Act, copyright subsists in the author only if the author ‘is a natural person, a human being, and not an artificial person’32 and that for the purposes of registration of a copyright, details of only a natural person may be provided as the author of a work. The ‘tool’ used for generating intellectual property is considered only as a machine (i.e. artificial person) and thus is not considered the ‘owner’ of such intellectual property.
However, in November 2020, the Copyright Office (established under the Copyright Act) for the first time recognised co-authorship of an artificial intelligence tool called ‘Raghav Artificial Intelligence Painting App’, and the owner of the said application in a painting titled ‘Suryast’.33
In this regard, the Standing Committee on IPR Regime in its report recommended ‘that a separate category of rights for AI and AI related inventions and solutions should be created for their protection as IPRs. It further recommended that the Department Related Parliamentary Standing Committee on Commerce should make efforts in reviewing the existing legislations of The Patents Act, 1970 and Copyright Act, 1957 to incorporate the emerging technologies of AI and AI related inventions in their ambit.’34
2.3. Categorisation and ownership of BD
2.3.1. BD is used by AI applications to discern a pattern from chaos using ML. It can be classified into two categories:
a. Personal Data – Governed and protected under ‘Right to Life’, a fundamental right granted under the Constitution of India, and provisions of the Information Technology Act, 2000 (‘Information Technology Act’).
b. Data other than personal data – presently not governed by any specific legislation.
2.3.2. India presently does not have any specific legislation governing data protection or privacy. However, in Justice K.S. Puttaswamy and Ors. v. Union of India (UOI) and Ors.,35 the Supreme Court of India read the ‘right to privacy’ into the other existing fundamental rights.
The data of natural persons (i.e. personal data) is protected under the right to privacy and the individual is the owner of such data. Such data when anonymised ceases to be ‘personal data’ and is available for analysis as far as creativity allows. Compilation of the anonymised data and interpretation thereof, may be protected under the copyright law.
2.4. Recognising the advances being made in the field of AI, it is imperative to have a data repository where anonymised personal data and other relevant data can be stored, pooled together, and made accessible based on identified parameters. It would seem that BD merits the status of ‘critical infrastructure’, allowing others to build upon the existing data, in a healthy competitive environment.
NITI Aayog is in the process of launching a program to develop a national repository of annotated and curated pathology images.36 The policy also talks about having a decentralised data marketplace that is based on blockchain technology, which will attract data providers and model builders to build AI applications.
3. Antitrust/competition laws
3.1. The Competition Act, 2002 (‘Competition Act’) seeks to prevent any adverse effect on competition, promote and sustain competition in markets, protect the interests of consumers, and ensure freedom of trade carried on by other participants in the market, in India.37 The Competition Act, amongst other things, prohibits: a. the abuse of a dominant position;
b. anti-competitive combinations; and
c. anti-competitive agreements.
3.2. The Advent of AI and the increasing use of BD have evidently accorded a dominant position to certain enterprises by allowing them to analyse and predict customer behaviour patterns and also develop cost efficiencies.
From a competition law perspective, an enterprise is said to have a dominant position in the relevant market38 if it enjoys ‘a position of strength which enables it to: (i) operate independently of competitive forces prevailing in the relevant market; or (ii) affect its competitors or consumers or the relevant market in its favour’.39 While enjoying such a position is not prohibited in itself, certain enterprises have been alleged to breach the Competition Act and abused their position40 in the market.
As an example, the Competition Commission of India (CCI)41 has noted that Amazon and Flipkart have ‘large repositories of data due to its unparalleled market base and market power’ and they analyse the data to ‘target advertisements based on consumer preferences and marginalise other competitors which are unable to capture the market due to lack of access to data’. Lack of such access and cost associated with the development of complex self-learning computing algorithms has resulted in the creation of high-entry barriers on account of network effects.
Although e-commerce entities essentially follow a marketplace model of e-commerce, i.e., acting as an online intermediary between sellers and consumers, they have now introduced private labels which are claimed to be given preferential treatment. It has also been alleged that e-commerce entities use the data collected from the sale of products of third-party sellers on its marketplace to set optimal prices and specification(s) for their private label products.
The CCI, by its order dated 20th January 2020 in Delhi Vyapar Mahasangh v. Flipkart Internet Private Limited and Amazon Seller Services Private Limited,42 has initiated an investigation focused on deep discounting, preferential listing and market power, but it still remains to be seen how the CCI plans to combat such behaviour.
3.3. The CCI has been further empowered to regulate mergers, acquisitions and amalgamations over a monetary threshold and prohibits any ‘combination which causes or is likely to cause an appreciable adverse effect on competition within the relevant market in India’.43
Although the quantum of data which could fall under the control of a single entity has not been a consideration of the CCI while approving a proposed combination, such non-price factors may result in repudiation of a proposed combination, keeping in mind ‘the extent of barriers to entry’44 it might create, and the ‘likelihood that the combination would result in removal of a vigorous and effective competitor or competitors in the market’45 among other factors.
3.4. Competing enterprises have sometimes resorted to coordinating their production and pricing activities to mimic a monopoly, for increasing their collective and individual profits by restricting market output and raising the market price. In response to such explicit or tacit collusion, the Competition Act prohibits anti-competitive agreements.46 Anti-competitive agreements include any agreement that ‘directly or indirectly determines purchase or sale prices47 or directly or indirectly results in bid rigging or collusive bidding’.48
The law as it stands does not provide for the use of AI applications as a means of collusion among competitors, but it is plausible that such arrangements may be deemed to be anticompetitive. For instance, if two or more enterprises, instead of agreeing on an explicit price, agree to implement a joint pricing algorithm that coordinates prices on their behalf.
3.6 The legal framework related to competition law will need to evolve so that AI is not available as a shield for enterprises to engage in activities which are otherwise prohibited. For example, the use of AI-based pricing software by competing enterprises resulting in collusion among themselves by determining similar prices for their goods or services.
4. Board of directors/governance
The board of directors (Board) is responsible for the management of the affairs of the company, and implementation of corporate governance. We will not delve into the merits of the statement, given that the directors’ powers may be curtailed in accordance with the Companies Legislation.49 For the purpose of this chapter, it should be brought into focus that the Companies Legislation and other statutes50 entrust certain duties to the directors.51
A director should, amongst other things:
a. Act in good faith in order to promote the objects of the company for the benefit of its members as a whole, and in the best interests of the company, its employees, the shareholders, the community and the environment.
b. Exercise duties with due and reasonable care, skill and diligence, and exercise independent judgment.
c. Not become involved in a situation in which he or she may have a direct or indirect interest that conflicts, or possibly may conflict, with the interest of the company.
d. Not achieve or attempt to achieve any undue gain or advantage either to himself or to his relatives, partners, or associates, and if such director is found guilty of making any undue gain, he shall be liable to pay an amount equal to that gain to the company. e. Not assign his office.
The focus of the Board in the course of discharging its responsibilities is engaged in developing a corporate strategy which essentially endeavours to upgrade technology to gain competitive advantage, increase production and reduce the cost of labour. In a nutshell, more often than not, the aim of this is to maximise the profits of the company. The involvement of AI in respect of governance is not hard to imagine.
4.2. The role of AI at Board level
Involvement of AI in corporate governance may be viewed at two levels. First, utilisation of AI by the Board for its decision making, and secondly, utilisation of machine learning for substituting one or more directors or perhaps even the entire Board.
A director is entrusted with the duty of exercising independent judgment, acting in good faith, not involving themselves in conflict-of-interest situations, and not assigning his/her office. It would follow that AI may be used at Board level by the directors, as long as the ‘duties’ are faithfully discharged. Substitution of a director by an AI is presently not envisaged under the Companies Legislation. The existing Companies Legislation clearly stipulates that only ‘individuals’ shall be elected as directors.52
In other jurisdictions, AI has been used by companies to conduct day-to-day business. A Hong Kong-based venture capital firm suggested the use of VITAL – an ML program capable of making investment recommendations in the biotechnology sector – to the Board.53 VITAL has been made an observer sitting on the Board, and corroboration by VITAL in all investment decisions was made mandatory.54 Similarly, a California-based software company runs all its corporate decisions through an AI tool, which gives its recommendations.55
Thus, under the present legal framework in India, the Board may use AI application assessments in making an informed decision while discharging its duties. Accordingly, it needs to be ensured that algorithms in question answer the requirements of the law. AI tools can give the company a competitive advantage and can be used to complete tasks involving due diligence and other administrative work for the Board, but it cannot undertake the decisions of a director on the Board.56
4.3. Governance issues and liability of the Board
The Board is empowered to take all decisions in respect of the company, except to the extent curtailed by provisions of the Companies Legislation and by the constituent documents. AI may be used to analyse customer demographics, analyse internal communication of employees to filter company data from leakage, review online news to point out competitors of the company, track capital allocation, etc.57 In recent times, a practice is emerging wherein corporate houses are resorting to AI for taking not just their day-to-day but also crucial business decisions.58 It is due to the transformation of the concept of AI from ‘general purpose technology’ to ‘general solution technology’, which has helped corporates and societies in solving real life problems.59
However, the Board must be equipped to deal with the surmounting challenges presented by the use of AI and be ready to address the various issues, including legal concerns, and take appropriate measures in handling the implementation of AI and the arising governance issues. Such issues include: data privacy; cyber security; biased programming in AI; and lack of transparency in the functioning of AI.
In the course of exercise of powers by the directors, the duties cast on the directors as discussed above must be fulfilled. Certain protocols should be developed and followed to ensure that the legal requirements are not directly or indirectly compromised in the guise of using AI. Before the official launch of any AI application, the Board should undertake alpha and beta testing to reduce product failure risk.
Failure on the part of the Board to address the implications arising out of the use of AI may result in penal liability.60 In the evolving regulatory expectations, a lapse in governance can have serious implications involving reputational damage, fall in stock price, and legal actions. Further, where personal data is involved, directors can be made liable for the failure to provide safety measures for data protection under the Information Technology Act.61 Further, intermediaries such as mutual funds or asset management companies have to frame additional policies or report cyber-attacks as well as include measures that they have taken to counter and mitigate such risks.62
4.4. Risk management solutions for the Board
Risk management measures at the Board level include:
a. Establishing protocols and having in place an independent mechanism to ensure that the protocols are adhered to in the use of AI. The protocols should serve to ensure amongst other things, compliance with applicable law.
b. Have in place a review mechanism to ensure that the protocols are evaluated on a periodic basis.
c. A separate mechanism in respect of processing data.
5. Regulations/government intervention
5.1. The rapid advancements and extensive applications of AI and ML have triggered profound interest, making them issues of national relevance. There is an urgent need for the Government to consider the development, funding and widespread implications of AI. Currently, there are no specific laws in India that relate to AI, BD or ML.
The Government’s priority at this stage seems to be in the promotion of AI and its application. Its strategy is to ‘maximize the late mover’s advantage’ in the AI sector for ‘consistently delivering home-grown pioneering technology solutions in AI as per its unique needs to help leap-frogging and catch-up with the rest of the world’.63
The Government is speeding up the process for formulation of laws, guidelines and policies, specifically for governing and regulating AI, BD and ML.
5.1.1. NITI Aayog Report. The Report suggests:
a. Building an attractive IP regime for AI innovation and recommends setting up a task force jointly comprising the Ministry of Corporate Affairs and the Department of Industrial Policy and Promotion (DIPP), to examine and issue appropriate modifications to the intellectual property laws.
b. Instituting a data privacy legal network to protect human rights and privacy and create sectoral regulatory guidelines covering privacy, security and ethics.
5.1.2. MeitY constituted four committees64 for developing a policy framework on AI. The recommendations made by the said committees include:
a. Development of an Open National Artificial Intelligence Resource Platform (NAIRP) to become the central hub for knowledge integration and dissemination in AI and ML.
b. Stakeholders must deliberate on whether AI systems should be recognised as a legal person and establish an insurance scheme or compensation fund to compensate for damages in the event of a civil liability claim.
c. Sharing of best practices – use of procurement contracts by the government to emphasise on the best practices around security, privacy and other issues.
d. A committee of the stakeholders should be constituted to look into the aspects in a holistic manner. Review of the existing laws to understand the modifications required for adoption of AI applications.
e. An AI framework should define broad principles and the organisations should be allowed to design their internal programmes in compliance with the set principles with flexibility to adapt to the developing technology.
f. Standards are to be set to address the AI development cycle. The Bureau of Indian Standards (BIS) has set up a new committee for standardisation in AI.
g. The Government has proposed the development of rigorous safety parameters and setting up of safety thresholds so that each AI application is designed ‘in such a way that it does not harm the people and property during its interaction’.65
5.1.3. In the AIRAWAT approach paper, NITI Aayog had proposed:
a. Setting up a specialised AI-computing infrastructure which will power the computing needs of Centres of Research Excellence, International Centres, Transformational AI and
Innovation Hubs, start-ups, researchers, students, and government organisations, etc.66
b. Setting up an inter-ministerial task force with cross-sectoral representation to spearhead the implementation of AIRAWAT.
c. The task force will seek funding for the implementation of AIRAWAT.
5.2. One of the committee reports67 has deliberated on whether AI poses a threat to humanity. It was opined that in the current state, AI applications are intelligent machines for specific tasks only. It stated that ‘even if a machine with higher intelligence is developed, there is no reason to believe that it would be interested in dominating the world due to lack of intent’.68 If machines with higher intelligence are developed, the ways of controlling those machines should also be developed in parallel.
5.3. Responsible AI. In 2020, NITI Aayog,69 policy think tank of the Government of India, issued two draft documents for the purpose of discussion with stakeholders on the subject of establishment of an oversight body and enforcement of responsible AI principles. The drafts proposed the following:
5.3.1. Establishment of an oversight body which shall play an enabling role under the following broad areas:
a. Managing and updating principles for responsible AI in India.
b. Researching the technical, legal, policy, and societal issues that come with AI.
c. Providing clarity on responsible behaviour through design structures, standards, guidelines, etc.
d. Enabling access to responsible AI tools and techniques.
e. Education and awareness on responsible AI.
f. Coordinating with various sectoral AI regulators to identify gaps and harmonise policies across sectors.
g. Representing India (and other emerging economies) in international AI dialogue on responsible AI.
5.3.2. It was further suggested that the oversight body shall be a highly participatory advisory body which shall interface with existing regulators across sectors and give dedicated resources to drive each mandate. It was stated that technology easily blends across other technologies and must not be viewed in silo.
Although the drafts are neither exhaustive nor final, they do reflect the intention of the government with respect to manner of regulation of AI and its use by private and public entities.
6.1. As is evident from the foregoing, there is an urgent need to develop a legal regime specific to AI, ML and BD. The following suggestions are made on the basis that the role of AI is set to become even more profound at all levels:
a. Develop a data repository where anonymised personal data and other relevant data can be stored, pooled together, and made accessible based on identified parameters.
b. Review of existing laws. Existing laws must be reviewed to keep up with the advancement in the field of AI and its diverse applications.
c. Enactment of special legislation related to AI, ML and BD.
i. The AI legislative framework should define broad parameters for the various stakeholders which include developers and users, and should have enough room to evolve based on stakeholder requirements. Such legislative framework should be enabling in character and allow innovation. Each stakeholder group should be required to design their internal programs and protocols in view of the legislative framework. ii. With a view to encourage innovation, the regulatory regime should have in place a mechanism to distinguish between error of judgment (where the person is innocent) and error of intent (where there is an element of mens rea).
6.2. Considering that the advancements in AI are occurring at a global level, India must seek to enhance its participation in various AI projects in co-ordination with international bodies. A good example is the close working arrangement of the WHO with the Ministry of Health and Family Welfare, Government of India to provide technical support for AI initiatives pursuant to its commitment to end tuberculosis by 2025.
* * *
1. PwC, ‘Sizing the prize: What’s the real value of AI for your business and how can you capitalise?’ (March, 2017). (Available at: https://www.pwc.com/gx/en/issues/ analytics/assets/pwc-ai-analysis-sizing-the-prize-report.pdf.)
2. Klaus Schwab, ‘The Fourth Industrial Revolution: What It Means and How To Respond’, Foreign Affairs, December 12, 2015. (Available at: https://www.foreignaffairs.com/ articles/2015-12-12/fourth-industrial-revolution.)
3. Brinda Sapra, ‘AI for All: How India Can Become an Artificial Intelligence Superpower’ Next Billion, November 1, 2019. (Available at: https://nextbillion.net/india-artificialintelligence-superpower/.)
4. Shashi Shekhar Vempati, ‘India and the Artificial Intelligence Revolution’, Carnegie India, August 11, 2016. (Available at: https://carnegieindia.org/2016/08/11/india-andartificial-intelligence-revolution-pub-64299.)
5. Manufacturing, fintech, healthcare, agriculture, education, retail, aid for differently abled persons, the environment, national security and public utility services.
6. NITI Aayog is the premier policy ‘think tank’ of India’s Government, providing both directional and policy inputs. While designing strategic and long-term policies and programmes for the Government of India, NITI Aayog also provides relevant technical advice to the Centre and States. (Available at: https://niti.gov.in/.)
7. Ministry of Electronics and Information Technology, ‘Committee A on platforms and data on Artificial Intelligence’, ‘Committee B on leveraging AI for identifying national missions in key sectors’, ‘Committee C on mapping technological capabilities, key policy enablers required across sectors, skilling, reskill’, ‘Committee D on cyber security, safety, legal and ethical issues’ (2018).
8. AI Research, Analytics and Knowledge Assimilation.
9. NITI Aayog, ‘AIRAWAT-establishing an AI specific cloud computing infrastructure for India’ (January, 2020). (Available at: https://niti.gov.in/sites/default/files/2020-01/ AIRAWAT_Approach_Paper.pdf.)
10. NITI Aayog, ‘Responsible AI for All’ 41 (2021). (Available at: https://www.niti.gov.in/ sites/default/files/2021-02/Responsible-AI-22022021.pdf.)
11. Rahul Sachitanand, ‘Here’s why Indian Companies are betting big on AI’, February 10, 2019, Economic Times. (Available at: https://economictimes.indiatimes.com/tech/ internet/heres-why-indian-companies-are-betting-big-on-ai/articleshow/67919349. cms?from=mdr.)
12. Press Information Bureau, Government of India, ‘NITI Aayog to Collaborate with IBM to develop Precision Agriculture using Artificial Intelligence’ (May 4, 2018).
13. Arindrajit Basu, ‘We need a better AI vision’, The Centre for Internet Society, October 12, 2019. (Available at: https://cis-india.org/internet-governance/blog/fountain-inkoctober-12-2019-arindrajit-basu-we-need-a-better-ai-vision.)
14. Sumit Das, Aritra Dey, Akah Pal, Nabamita Roy, ‘Applications of Artificial Intelligence in Machine Learning: Review and Prospect’, 115-No.9 International Journal of Computer Applications 31–32 (2015).
15. Press Trust of India, ‘India ranks 145 of 195 countries in healthcare access and quality, far below China’, Hindustan Times, May 23, 2018. (Available at: https://www. hindustantimes.com/health/india-ranks-145th-below-china-bangladesh-among-195countries-in-healthcare-access-quality/story-31UgnP7QxpvqbeHJoLdtFP.html.)
16. Microsoft News Center India, ‘Forus Health democratizes eye care with an ‘Intelligent Edge’ in its retinal imaging devices’, 6th April 2018. (Available at: https://news.microsoft. com/en-in/features/forus-health-3nethra-ai-azure-iot-intelligent-edge-eyecare/.)
17. Microsoft News Center India, ‘Forus Health democratizes eye care with an ‘Intelligent Edge’ in its retinal imaging devices’, 6th April 2018. (Available at: https://news.microsoft. com/en-in/features/forus-health-3nethra-ai-azure-iot-intelligent-edge-eyecare/.)
18. Microsoft News Center India, ‘Forus Health democratizes eye care with an ‘Intelligent Edge’ in its retinal imaging devices’, 6th April 2018. (Available at: https://news.microsoft. com/en-in/features/forus-health-3nethra-ai-azure-iot-intelligent-edge-eyecare/.)
19. NITI Aayog. (Available at: https://indiaai.gov.in/ministries/niti-aayog (last visited on January 31, 2022).)
20. Indian Brand Equity Foundation, ‘Indian Ecommerce Industry Report’ January 2020. (Available at: https://www.ibef.org/industry/ecommerce.aspx.)
21. Maj. Gen P.K. Chakravorty, ‘Artificial Intelligence and Its Impact on the Indian Armed Forces’, Indian Defence Review 5th May, 2017. (Available at: http://www. indiandefencereview.com/news/artificial-intelligence-and-its-impact-on-the-indianarmed-forces/.)
22. PwC, ‘AI: An opportunity amidst a crisis’ (December, 2020). (Available at: https:// www.pwc.in/assets/pdfs/data-and-analytics/ai-an-opportunity-amidst-a-crisis.pdf.)
23. Government of India, ‘75 @ 75: India’s AI Journey’ (Ministry of Electronics and Information Technology, 2021).
24. A New AI-Driven Platform will Facilitate Early-COVID Interventions over WhatsApp. (Available at: https://dst.gov.in/new-ai-driven-platform-will-facilitate-early-covid-inter ventions-over-whatsapp (last visited on January 31, 2022).)
25. A New AI-Driven Platform will Facilitate Early-COVID Interventions over WhatsApp. (Available at: https://dst.gov.in/new-ai-driven-platform-will-facilitate-early-covid-inter ventions-over-whatsapp (last visited on January 31, 2022).)
26. ‘IIT Jodhpur Researchers Develop COVID-19 Diagnosis Technique Using Chest X-Ray Images’ The Economic Times, January 28, 2022. (Available at: https://economictimes. indiatimes.com/news/science/iit-jodhpur-researchers-develop-covid-19-diagnosistechnique-using-chest-x-ray-images/articleshow/89179373.cms; https://timesofindia.ind iatimes.com/india/iit-jodhpur-team-develops-covid-diagnosis-technique-using-chest-xray/articleshow/89180305.cms.)
27. ‘IIT Jodhpur Researchers Develop COVID-19 Diagnosis Technique Using Chest X-Ray Images’ The Economic Times, January 28, 2022. (Available at: https://economictimes. indiatimes.com/news/science/iit-jodhpur-researchers-develop-covid-19-diagnosistechnique-using-chest-x-ray-images/articleshow/89179373.cms; https://timesofindia.ind iatimes.com/india/iit-jodhpur-team-develops-covid-diagnosis-technique-using-chest-xray/articleshow/89180305.cms.)
28. ‘IIT Jodhpur Researchers Develop COVID-19 Diagnosis Technique Using Chest X-Ray Images’ The Economic Times, January 28, 2022. (Available at: https://economictimes. indiatimes.com/news/science/iit-jodhpur-researchers-develop-covid-19-diagnosistechnique-using-chest-x-ray-images/articleshow/89179373.cms; https://timesofindia.ind iatimes.com/india/iit-jodhpur-team-develops-covid-diagnosis-technique-using-chest-xray/articleshow/89180305.cms.)
29. Anivar A. Aravind v. Ministry of Home Affairs & Ors. (W.P. No.7483 of 2020).
30. https://rajyasabha.nic.in/rsnew/Committee_site/Committee_File/ReportFile/13/141/ 161_2021_7_15.pdf.
31. The Copyright Act, 1957 (14 of 1957), s. 2(j).
32. Rupendra Kashyap v. Jiwan Publishing House Pvt. Ltd (1996(38) DRJ 81).
33. India Recognises AI as Author of a Copyrighted Work. (Available at: https://www.
lexcampus.in/india-recognises-ai-as-author-of-a-copyrighted-work/ (last visited on 31st January 2022).)
34. Government of India, ‘Review of the Intellectual Property Rights Regime in India’ (Department Related Parliamentary Standing Committee on Commerce, 2021).
35. AIR 2017 SC 4161.
36. NITI Aayog, ‘National Strategy for Artificial Intelligence’ (2018).
37. The Competition Act, 2002 (Act No. 12 of 2003), Preamble.
38. ‘Relevant Market’ means ‘the market which may be determined by the CCI with reference to the relevant product or geographic market or with reference to both; as per The Competition Act, 2002 (Act No. 12 of 2003), s. 2(r).
39. The Competition Act, 2002 (Act No. 12 of 2003), s. 4 (Explanation (a)).
40. The Competition Act, 2002 (Act No. 12 of 2003), s. 4(1).
41. Case No. 40 of 2019, Order dated January 13, 2020.
42. Case No. 40 of 2019, Order dated January 13, 2020.
43. The Competition Act, 2002 (Act No. 12 of 2003), s. 6(1).
44. The Competition Act, 2002 (Act No. 12 of 2003), s. 20(4)(b).
45. The Competition Act, 2002 (Act No. 12 of 2003), s. 20(4)(i).
46. The Competition Act, 2002 (Act No. 12 of 2003), s. 3(1), stipulates ‘no enterprise or association of enterprises or person or association of persons shall enter into any agreement in respect of production, supply, distribution, storage, acquisition or control of goods or provision of services, which causes or is likely to cause an appreciable adverse effect on competition within India’.
47. The Competition Act, 2002 (Act No. 12 of 2003), s. 3(3)(a).
48. The Competition Act, 2002 (Act No. 12 of 2003), s. 3(3)(d).
49. The Companies Act, 2013 (Act 18 of 2013).
50. Indian Trust Act, 1882 (Act 2 of 1882).
51. The Companies Act, 2013 (Act 18 of 2013), s. 166.
52. The Companies Act, 2013 (Act 18 of 2013), s. 149.
53. Nicky Burridge, ‘Artificial intelligence gets a seat in the boardroom’, Nikkei Asian Review, 10th May 2017. (Available at: https://asia.nikkei.com/Business/Artificialintelligence-gets-a-seat-in-the-boardroom.)
54. Nicky Burridge, ‘Artificial intelligence gets a seat in the boardroom’, Nikkei Asian Review, 10th May 2017. (Available at: https://asia.nikkei.com/Business/Artificial-intelligencegets-a-seat-in-the-boardroom.)
55. Martin Petrin, “Corporate Management in the age of AI” 3 UCL Working paper series (2019).
56. Nicky Burridge, ‘Artificial intelligence gets a seat in the boardroom’, Nikkei Asian Review, 10th May 2017. (Available at: https://asia.nikkei.com/Business/Artificialintelligence-gets-a-seat-in-the-boardroom.)
57. Erman Akdogan, ‘Blockchain as the Board, AI as the Director — Corporate Governance 2.0’, The Startup, 22nd November 2019. (Available at: https://medium.com/swlh/ blockchain-as-the-board-ai-as-the-director-corporate-governance-2-0-a07401350358.)
58. Welcome the Newest Member to Your Board of Directors: AI. (Available at: https:// www.govenda.com/blog/artificial-intelligence-in-corporate-governance/ (last visited on January 30, 2022).)
59. Michael Hilb, ‘Toward Artifcial Governance? The Role of Artifcial Intelligence in Shaping the Future of Corporate Governance’ 24 Journal of Management and Governance 851 (2020). (Available at: https://link.springer.com/content/pdf/10.1007/ s10997-020-09519-9.pdf.)
60. The Companies Act, 2013 (Act 18 of 2013), s. 166 (7).
61. The Information technology Act, 2000 (Act 21 of 2000), s. 85.
62. SEBI Circular No. SEBI/HO/IMD/MIRSD/CIR/P/2019/12 and SEBI/HO/IMD/ MIRSD/CIR/P/2018/147.
63. NITI Aayog, ‘National Strategy for Artificial Intelligence’ (2018).
64. Ministry of Electronics and Information Technology, ‘Committee A on platforms and data on Artificial Intelligence’, ‘Committee B on leveraging A.I. for identifying national missions in key sectors’, ‘Committee C on mapping technological capabilities, key policy enablers required across sectors, skilling, reskill’, and ‘Committee D on cyber security, safety, legal and ethical issues’ (2018).
65. Ministry of Electronics and Information Technology, ‘Committee D on cyber security, safety, legal and ethical issues’ (2018).
66. NITI Aayog, ‘AIRAWAT: Establishing an AI specific cloud computing infrastructure for India, An approach Paper’, (January 2020).
67. Ministry of Electronics and Information Technology, ‘Committee D on cyber security, safety, legal and ethical issues’ (2018).
68. Ministry of Electronics and Information Technology, ‘Committee D on cyber security, safety, legal and ethical issues’ (2018).
69. NITI Aayog, ‘Working Document: Enforcement Mechanisms for Responsible
#AIforAll Part 1’. (Available at: https://niti.gov.in/sites/default/files/2020-11/Towards_
Responsible_AIforAll_Part1.pdf) and NITI Aayog, ‘Working Document: Enforcement Mechanisms for Responsible #AIforAll Part 2’ (available at: https://niti.gov.in/sites/ default/files/2020-11/Towards-Responsible-AI-Enforcement-of-Principles.pdf).