Context: The Ministry of Electronics and Information Technology (MeitY) is exploring the idea to establish an AI Safety Institute (AISI) in India under the IndiaAI Mission.
Relevance of the Topic: Prelims: Key facts about AI Safety Institute (once it is established); Global Partnership on AI (GPAI) initiative; Bletchley Declaration; IndiaAI Mission.
Major Highlights:
- In recent years, India has displayed leadership in developing a robust Artificial Intelligence ecosystem and AI governance at the G20 and the Global Partnership on AI (GPAI) initiative.
- India needs to consider setting up its own globally assimilated, but locally driven, AI institute to create an ecosystem for cutting-edge AI innovation, access, and safety, as outlined in the Bletchley Declaration.
Bletchley Declaration:
- Bletchley Declaration is a global agreement on the responsible development of AI.
- Aim: Enhance global cooperation on artificial intelligence (AI) safety by identifying AI-related risks and developing collaborative policies for mitigation of these risks.
- It was signed by the 28 countries & EU at the AI Safety Summit, UK, in November 2023.
- Important signatories: India, China, the US, the UK, European Union etc.
What is AI Safety?
- AI safety refers to practices and principles that help ensure AI technologies are designed as responsibly as possible to benefit humanity and minimize any potential harm or negative outcomes. It includes:
- Ethical design of algorithms
- Ensuring data privacy and security of individuals and organisations
- Identifying potential AI risks (such as bias, data security, vulnerability to external threats) and developing AI safety measures for risk mitigation.
Scope of AI Safety Institute:
- Advocate responsible AI deployment adapting to the unique needs of industries such as healthcare, finance, and logistics etc.
- Facilitate proactive information sharing without being a regulator.
- Assess the risk to public safety from frontier AI models by leveraging multi-stakeholder consortiums and partnerships.
- Improve government capacity and mainstream the idea of external third-party testing and risk mitigation and assessment.
- Deliver insights which can transform AI governance into an evidence-based discipline.
Structure of AI Safety Institute:
- Standardisation Agency: AISI should be a technical institution that operates exclusively as a technical research, testing, and standardisation agency which sets standards for AI safety. It should be independent from rulemaking and enforcement authorities.
- Advisory role of AISI: AISI’s role should not be limited to safety testing and standard-setting but should also be advisory in nature, helping policymakers and the private sector understand and mitigate the socio-technical risks AI poses. The institute could champion perspectives on risks relating to bias, discrimination, social exclusion, gendered risks, labour markets, data collection and individual privacy.
- Multi-stakeholder collaboration: In collaboration with a broad range of stakeholders (including startups, large enterprises, academic institutions, civil society organizations, and government bodies), AISI should work closely to develop and disseminate industry best practices, responsible AI use guidelines, and advocate the importance of responsible AI practices across sectors.
- Scalability and Global Engagement: AISI should collaborate with governments and stakeholders from across the world. Shared expertise will be essential to keep up with AI’s rapid innovation trajectories and help in scaling the capabilities of AI.
The AI Safety Institute would help India become a global steward for forward-thinking AI governance which embraces many stakeholders and government collaboration. AISI can demonstrate India’s scientific temper and willingness to implement globally compatible, evidence-based and proportionate policy solutions.