India Announces AI Regulatory Advisory

Mar28

Posted by admin
Posted in Blog

The advisory represents a change in India’s regulatory strategy – moving from a lack of regulation to a more cautious approach regarding AI regulation.

India has recently made a significant entry into the global AI regulations debate by issuing an advisory. The Ministry of Electronics and IT in India released the advisory on March 1, 2024. Subsequently, an amended version was released on March 15. While the advisory is not legally binding, it signals a shift in India’s approach to AI regulation and sets the tone for future regulations in the country.

The advisory aims to ensure that tech firms’ products and services do not exhibit bias, discrimination, or threaten the integrity of the electoral process. The ministry cites the authority granted to it through the IT Act, 2000, and IT Rules, 2021 as the basis for the advisory.

Amended version less rigid in approach

On March 15, the Ministry issued an updated version of the March 1 advisory. It is interesting to note that this amended version has diluted the rigidity of approach that was evident in the previous release. While the majority of clauses from the earlier version have been retained in the New Advisory, there have been some changes to the regulatory framework. These are:

Removal of Prior Government Approval: Previously, intermediaries and platforms were obligated to obtain government approval before making under-testing or unreliable AI tools available to the Indian public. The updated version eliminates this approval requirement.

Emphasis on Labelling: The advisory highlights the importance of user awareness regarding AI-generated content. Intermediaries and platforms are now required to appropriately label AI, especially when it is under-tested or unreliable, to inform users about the potential fallibility of outputs. User awareness can be achieved through consent pop-ups or other relevant mechanisms. This transparency enables users to critically assess the information they encounter online and helps prevent manipulation through AI-powered misinformation campaigns. AI-generated content, particularly susceptible to misuse, such as deepfakes, must be clearly labelled to raise user awareness.

Removal of Reporting Requirement: In the first version, intermediaries and platforms were mandated to submit an Action Taken-cum-Status Report to the Ministry of Electronics and Information Technology within 15 days. However, the revised advisory eliminates this reporting obligation altogether.

Signal of the future of regulation

IT Deputy Minister Rajeev Chandrasekhar has stated that the advisory is a signal of the future of regulation. He clarified that the advisory targets “significant” platforms and does not apply to start-ups. Chandrasekhar also highlighted the importance of appropriate labelling of the fallibility or unreliability of AI model outputs.

This advisory represents a departure from India’s previous hands-off approach to AI regulation, where the sector was considered strategically important. The change has surprised many industry executives, including Indian startups, venture capitalists, and Silicon Valley leaders. They have expressed concerns about the impact of the regulation on India’s global competitiveness.

The advisory was prompted, in part, by dissatisfaction with Google’s response to a user query about India’s Prime Minister Narendra Modi. Google’s AI model, Gemini, provided a response that some considered to be in violation of the IT Rules, 2021. The non-compliance with these rules may result in penal consequences, according to the advisory.

The industry is cautious

The introduction of the advisory represents a change in India’s regulatory strategy, moving from a lack of regulation to a more cautious approach regarding AI regulation. While some industry experts and tech leaders have welcomed the advisory as a step toward responsible AI deployment, others have criticised it. Many Indian startups and venture capitalists are concerned that the regulation will hinder India’s ability to compete globally in the AI race. Some Silicon Valley leaders have also expressed disappointment with India’s policy shift. Although the approval mechanism has been eliminated, intermediaries and platforms still face considerable risks due to the lack of clarity and ambiguity in the standards for testing unreliable AI systems.

The advisory reflects an increased focus on ensuring the ethical and responsible use of AI technology in India. It remains to be seen how this advisory will shape the future of AI regulation in India and its impact on the country’s tech industry.

[To be concluded]