AI Risks and Governance: Balancing Productivity and Concentrated Power – Insights by Prof. Saumitra Bhaduri

Overview Professor Saumitra Bhaduri of the Madras School of Economics discusses how AI can boost productivity but also creates serious risks that need careful governance. Key Developments AI is moving from automating physical work to disrupting cognitive labour , raising concerns about job displacement. Development of advanced frontier models is highly capital‑intensive , concentrating ownership in a handful of corporations and nations. The rise of synthetic information creates new threats of misinformation and manipulation. Calls for a global non‑proliferation agreement on disruptive AI and a robust governance architecture to ensure accountability. Important Facts 1. AI can enhance productivity across sectors if deployed responsibly. 2. Historical patterns show technology reshapes, not eliminates, work; reskilling is essential. 3. Concentrated AI capability can become a strategic chokepoint, influencing geopolitics. 4. AI lacks true knowledge; it predicts patterns without understanding context. 5. Organizations increasingly rely on AI for research, coding, legal drafting, and finance, raising systemic risk. UPSC Relevance The discussion touches on multiple GS papers. Understanding AI is vital for GS3 (Technology & Economy) and GS4 (Ethics). The impact on cognitive labour relates to employment trends, a key GS3 topic. Concentration of AI power and the need for international treaties link to GS2 (Polity) and global governance. Ethical concerns about synthetic information are relevant for GS4. Way Forward Invest in large‑scale reskilling programs to help workers transition to AI‑augmented roles. Promote transparent AI development through open standards and audit mechanisms. Negotiate an international non‑proliferation agreement to limit the spread of high‑risk AI systems. Establish clear liability frameworks so that AI providers can be held accountable for harms. Encourage interdisciplinary education that combines technical skills with ethical judgment, making the ability to verify AI outputs a prized expertise.