India to Establish AI Safety Institute: Key Points

Published Date::13/10/2024

The Indian government is considering the establishment of an Artificial Intelligence Safety Institute (AISI) to set standards and guidelines for AI development without stifling innovation. This move follows similar initiatives by the UK, USA, and Japan, and aims to foster a safe and collaborative AI ecosystem.

The Indian government is mulling the creation of an Artificial Intelligence Safety Institute (AISI) to establish standards, frameworks, and guidelines for AI development. This initiative, led by the Union Ministry of Electronics and Information Technology (MeitY), seeks to promote AI safety without acting as a regulatory body or stifling innovation. The decision comes after detailed consultations with stakeholders on October 7, according to sources familiar with the matter.

 Global ContextThe UK was the first to announce an AISI during the AI Safety Summit held in Bletchley Park in November 2023, with an initial investment of £100 million (approximately ₹1,100 crore). The USA followed suit by setting up its AISI as part of the National Institute of Standards and Technology (NIST), and Japan launched its AISI in February 2024.

In May 2024, the European Union and eleven other nations, including the USA, UK, South Korea, Australia, Canada, France, Germany, Italy, Japan, South Korea, and Singapore, signed the Seoul Declaration at the Seoul Summit. The declaration aimed to create or expand AISIs, research programs, and institutions to promote cooperation on safety research and share best practices.

 Stakeholder EngagementAhead of the October 7 meeting, MeitY shared two categories of questions with stakeholders. These included companies such as Meta, Google, Microsoft, IBM, and OpenAI; industry bodies like Nasscom, Broadband India Forum, and BSA-The Software Alliance; multiple IITs; consulting firms such as The Quantum Hub and Dialogue; and civil society organizations like the Digital Empowerment Foundation and Access Now.

The first category of questions focused on AISI's core objectives, its optimal organizational structure, scalability, development of indigenous AI safety tools tailored to India's unique challenges, and potential strategic partners. The second category delved into strategies for stakeholder engagement, national and international partnerships, and the AISI's role in global AI safety discussions and standards.

 Key Consensus PointsParticipants in the consultation meeting reached several key consensus points. First, there is a strong belief that the government is serious about setting up AISI. Second, AISI is not intended to be a regulatory body but will focus on identifying potential harms and setting standards. Third, interoperable systems are essential to prevent silos in AI development.

The AISI could develop a risk assessment toolkit and a voluntary compliance toolkit for the industry, although its frameworks will not be binding. Research supported by AISI could inform future AI policies drafted by the Indian government.

 Structure and IndependenceStakeholders discussed various structural possibilities for the Indian AISI, including whether it should be housed within MeitY or within academic institutions like IITs to ensure reasonable independence. There was also a discussion on the role states could play, especially those like Telangana that are already using AI in public service delivery. The AISI could provide guidance and best practices without being prescriptive.

 Budget and FundingDuring the October 7 meeting, there was a short discussion on budget allocation for the Indian AISI. Abhishek Singh, who oversees India AI, mentioned that about ₹20 crore has been earmarked for AISI under the IndiaAI Mission, with the possibility of additional funds from other verticals of the mission. The cabinet had approved a budget outlay of ₹10,372 crore over five years for the IndiaAI Mission in March, and in the FY25 budget, the government allocated ₹551.75 crore for the mission.

 Mandate and FlexibilityStakeholders agreed that the AISI should have the budget and flexibility to evolve both its structure and mandate. Initially, the mandate could be limited to setting standards for AI, understanding risks associated with its development and deployment, and fostering international collaborations. A broad agenda could be overwhelming and counterproductive.

The establishment of AISI is a significant step towards ensuring that AI development in India is safe, ethical, and aligned with global best practices. It reflects the government's commitment to fostering innovation while addressing potential risks and challenges in the AI space.

FAQS:

Q: What is the main purpose of the Artificial Intelligence Safety Institute (AISI) in India?

A: The main purpose of the AISI in India is to set standards, frameworks, and guidelines for AI development, promoting safety and ethical practices without stifling innovation.

Q: How does AISI differ from regulatory bodies?

A: AISI is not a regulatory body. It focuses on identifying potential harms, setting standards, and providing guidelines, but its frameworks will not be binding.

Q: Which countries have already established their own AISIs?

A: The UK, USA, and Japan have already established their own AISIs. The UK was the first to announce its AISI during the AI Safety Summit in November 2023.

Q: What role do stakeholders play in the AISI's development?

A: Stakeholders, including companies, industry bodies, academic institutions, and civil society organizations, provide inputs on the AISI's structure, mandate, and strategic partnerships to ensure it meets the needs of the AI ecosystem.

Q: How much funding has been allocated for the Indian AISI?

A: About ₹20 crore has been earmarked for the AISI under the IndiaAI Mission, with the possibility of additional funds from other verticals of the mission.

More Topics :