Exploring a Consent-Based Framework for AI Regulation
The Ministry of Electronics and Information Technology (MeitY) is considering a new regulatory framework for artificial intelligence (AI) in India. This framework aims to address issues around synthetically-generated content, agentic AI autonomy, and regulatory sandboxes for high-risk AI applications.
Synthetically-Generated Content
- Synthetically-generated content refers to digital media created or altered by AI rather than humans.
- MeitY is proposing a consent-based framework for such content to address concerns about its proliferation, particularly in the form of deepfakes.
Agentic AI Autonomy
- Agentic AI refers to systems with autonomy to plan and execute multi-step actions with minimal human supervision.
- The government is considering regulations on the autonomy and data retention capabilities of such AI systems.
Regulatory Sandboxes
- MeitY plans to consult RBI and SEBI to create regulatory sandboxes for AI applications, focusing on finance and public services.
AI Liability Concerns
- The liability of AI platforms and developers for content generated by AI models is a major discussion point.
- Unlike social media platforms that enjoy safe harbor, AI models' liability is complex due to their editorial nature and training on internet content.
Data Protection and AI
- The Digital Personal Data Protection Act, 2023, addresses privacy issues but raises questions about consent and data ownership in AI.
- Data voluntarily made public by individuals is exempt from protection under the data protection law, posing challenges for AI content scraping.
Current Legislative Efforts
- Though India initially believed existing laws sufficed for AI regulation, recent developments prompted a reconsideration.
- Past measures include requiring social media platforms to label AI-generated content continuously.
International Influences
- India is examining Singapore’s governance framework for agentic AI as a potential model.
