Pvt member’s Bill to regulate AI introduced in Rajya Sabha

New Delhi: Communist Party of India (CPI) Rajya Sabha MP P Sandosh Kumar introduced a Private Member’s Bill on Friday to regulate artificial intelligence (AI) and deepfakes by establishing a National Artificial Intelligence Regulatory Authority (NAIRA).

The Bill proposes to establish NAIRA as the apex authority for the regulation and monitoring of AI deployment and related technologies in the country, as well as assessing risks (including threats to employment). HT has seen a copy of the Bill.

Kumar proposed that this authority should consist of a chairperson and two members, appointed by the central government for a period of at most four years, who have experience in the field of AI.

The two members should understand the risks associated with AI, machine learning, and deep learning and should have worked with organisations that focused on ethically deploying AI and understand the “future threats of AI in replacing human jobs.”

This proposal comes days after the Economic Survey noted that the social impact of emerging technologies such as AI via labour market disruptions and labour displacement “is barely understood.” The survey said that AI causes “a huge pall of uncertainty” over workers of all skill levels. It also noted the immense energy demand of AI and called it “an energy guzzler.”

In the proposed Bill, Kumar has defined ‘artificial intelligence’ as “a constellation of technologies that enable machines to act with higher levels of intelligence and emulate the human capabilities of sense, comprehend, and act.” He has defined ‘deep learning’ as a method in AI that “teaches computers to process data in a way that is inspired by the human brain.” ‘Machine learning’ has been defined as a field of study of AI “related to the development and study of statistical algorithms that can learn from data and generalise to unseen data and thus perform tasks without explicit instructions.”

Kumar has also defined ‘deep fake’ as artificial media that has been digitally altered to replace a person’s face or body with that of another.

The bill proposes that NAIRA should be responsible for developing the monitoring and regulatory landscape for the adaptation of AI and to identify key privacy challenges posed by the technology. It also wants NAIRA to deal with complaints related to “the deprivation of rights of workers” due to the replacement of human labour with AI and to act as a grievance redress mechanism.

The government, through its nodal ministry—the ministry of electronics and information technology—has been contemplating regulating AI and deepfakes. For regulation of deepfakes, earlier this year, there was discussion around amending the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021, that gained significant momentum before the announcement of elections, but it petered out after deepfakes barely surfaced during the general elections.

For AI, the government had released an advisory about needing the government’s permission to deploy AI models. This was later rescinded and replaced with a watered-down advisory that did not require the government’s approval.

There is recognition within the ministry of electronics and information technology that a full-fledged law to deal with challenges posed by AI may take a long time, and it is discussing how to put in checks and balances through either rules or policy, HT has learnt. Nothing has been finalised yet, given the evolving nature of the field. MeitY also does not want to stifle innovation in the sector by prematurely regulating it.