Artificial Intelligence (AI) is rapidly changing our world, from improving healthcare to enhancing transportation systems. However, with the increasing impact of AI, the regulation of its development and use has become a major concern for governments and international organizations. The regulation of AI is essential to ensure that the technology is developed and used in a responsible and ethical manner. In recent years, global trends in the regulation of AI have been emerging, with different countries and regions taking different approaches to address the challenges posed by this technology.
The Indian government sees AI as a "kinetic enabler" and seeks to harness its potential for better governance. As per Ministry of Electronics and Information Technology ("MeitY"), AI can be used to provide personalized and interactive citizen-centric services through Digital Public Platforms. 1 The Indian government is not considering bringing a law governing AI as of now.2 This, however, does not mean no regulation at all. Regulation of AI presently is undertaken via regulations on issues like intellectual property, privacy and cyber security. The Indian government has also undertaken several other initiatives to enable the AI ecosystem in the country to thrive. In 2018, NITI Aayog published its discussion paper titled National Strategy for Artificial Intelligence identifying healthcare, education, agriculture, smart cities and mobility as some of the sectors for deployment of AI. MeitY has also operationalized a clutch of Centers of Excellences (COEs) to assist in knowledge management and creating capabilities to capture new and emerging areas of technology.
There have been attempts made by other regulators to initiate a discussion on the utilization of these technologies. For instance, Securities and Exchange Board of India (SEBI) has proposed to use AI for data analytics and pattern recognition to aid the regulator in identifying fraudulent behaviour and abnormal trading. The Reserve Bank of India (RBI) has also given an expression of interest for engaging consultants in the use of AI and machine learning, and advanced analytics.
In recent discussions on the Digital India Act ("DIA"), which is being created as a comprehensive legal framework for the digital ecosystem, MeitY stated that the proposed law would regulate "high risk AI systems" through legal, quality testing framework to examine regulatory models, algorithmic accountability, zero-day threat & vulnerability assessment, examining AI based ad-targeting, content moderation, etc. However, this proposed law does not specifically address the challenges created by use of AI.
The European Union proposes to create a comprehensive and substantial legal framework to govern AI. The EU's proposed Artificial Intelligence Act ("AIA") contemplates a graded regulation, i.e., regulation of AI based on its potential risks. The AI systems that are perceived as a threat to safety, livelihood of people and human rights are prohibited under this framework. They are placed in the "unacceptable risk" tier. This includes AI processes that have the potential to manipulate people, or that which may exploit vulnerabilities of specific groups, such as persons with disabilities. The AI systems which are not prohibited are categorized into "high-risk", "limited risk" and "minimal risk". The AIA is subject to EU's legislative process and may be amended in the coming years.
Australia is another jurisdiction that is grappling with regulatory issues surrounding AI. While there is no legislation which specifically deals with AI in Australia, the Privacy Act Review Report 2022 ("Review Report"), if adopted, may have some implications on the use of AI. The Review Report recommends that entities be required to disclose how personal information would be used in automated decisions. The black box effect in AI systems makes it difficult to regulate automated decisions because the decision-making process is not transparent, and the rationale behind the decisions is not always clear. This lack of transparency and interpretability can create ethical, legal, and regulatory challenges, particularly when it comes to ensuring fairness and preventing discrimination in decision-making.
AI and machine learning are increasingly being used for individual profiling, based on data such as behavioral trends, location, education, and purchasing history. This profiling can be used to make decisions about product and service offerings, but can also lead to bias in areas such as recruitment, and can be used for surveillance purposes. Additionally, non-consented use of AI for aggregating data and extracting health information can unfairly influence opinions, choices, and offers made to individuals by businesses, and the opaque nature of AI algorithms can make it difficult to determine how data is being used. These issues highlight the need for regulation and ethical considerations in the use of AI for individual profiling and other purposes.
Governments and legal systems worldwide are struggling to deal with the challenges posed by the black box effect and the use of AI in decision-making, particularly when it comes to protecting privacy rights and ensuring fairness and accountability.
Amidst these debates on regulation of AI, Elon Musk and others have been voicing concerns over the rapid development of AI, highlighting the dangers it poses to human civilization. There is a growing fear that AI may eventually outnumber and outsmart human minds. An open letter endorsed by the likes of Elon Musk, Steve Wozniak and Yuval Harari has suggested a pause on training systems more powerful than GPT-4 so that AI is made more transparent, safe and loyal.
Footnotes
1. "Generative Artificial Intelligence", Press Information Bureau, Ministry of Electronics and Information Technology, (February 3, 2023) available at https://pib.gov.in/PressReleasePage.aspx?PRID=1896016.
2. "Government Not Considering Regulating AI Growth, Says IT Minister Vaishnaw", Economic Times (April 5, 2023) available at https://economictimes.indiatimes.com/news/india/government-not-considering-regulating-ai-growth-says-it-minister-vaishnaw/articleshow/99273629.cms?from=mdr.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
