The White Paper released by the Office of the Principal Scientific Adviser (OPSA) proposes a shift from traditional "command-and-control" regulations to a “techno-legal model” for AI governance.

Key principles of the techno-legal framework 

• Governance by Design: Legal and ethical requirements are embedded into AI systems at the design stage.
• Pro-Innovation and Flexible Regulation: Risk mitigation is balanced with innovation.
• Risk-Proportionate Controls: The intensity of governance varies according to the scale and potential harm of AI deployment.
• Human Oversight: Human supervision is retained at critical decision points to prevent automated harm.
• Lifecycle-Based Governance: Safeguards operate across the entire AI lifecycle; from data collection and training to deployment and use (refer infographic). 

Challenges in Implementation

• Privacy vs. Performance: Large-scale data erasure can reduce model performance, especially for underrepresented linguistic and cultural groups.
• Distinction Between AI Users and AI Subjects: In Indian welfare contexts like healthcare and public safety, citizens are often subjects rather than users, lacking the ability to contest outcomes.
• Cross-Border Alignment and Enforcement: A model trained abroad may not incorporate Indian safeguards, creating a regulatory gap. 
• Balancing Compliance with System Accuracy: Introducing techno-legal measures can potentially affect the performance of AI systems.