Absolutely chuffed to hear that Daren Acemoglu (one of my fav econ writers), Simon Johnson and Robinson won the Nobel memorial prize. Acemoglu has been a brill

Absolutely chuffed to hear that Daren Acemoglu (one of my fav econ writers), Simon Johnson and Robinson won the Nobel memorial prize. Acemoglu has been a brilliant and prolific econ writer, and I enjoy his exposition. Many have commented on his "Why Nations Fail", and rightly so. More recently, he also wrote "Power and Progress: Our Thousand-Year Struggle Over Technology and Prosperity". I'm more interested in his thoughts on tech and development (than on instituions and democracy), and he doesn't disappoint. More topically for HN folks, he wrote on "The Simple Macroeconomics of AI" [1]. Unlike many contemporary researchers / writers (incl. Ethan Mollick) who are completely ga-ga over prospects of AI-driven economy over the next decade, Acemoglu is more reserved. His top findings: Modest Productivity Gains: Estimates suggest TFP growth from AI will be less than 0.55% to 0.71% over 10 years, limited by the complexity of tasks. Hard vs. Easy Tasks: AI's impact on "hard-to-learn" tasks, which lack clear metrics and are context-sensitive, will likely produce smaller gains than on simpler tasks. Wage Effects: AI's productivity benefits may not substantially raise wages, particularly for low-skill workers. Inequality Impact: AI is unlikely to reduce labor income inequality significantly and may widen the gap between capital and labor incomes. Task Complementarity: AI could enhance productivity by complementing human workers, though benefits may be limited if task prices fall. Role of New Tasks: New AI-enabled tasks could boost productivity, but some may also create negative social value, such as manipulative content. Automation Limitations: Extensive-margin automation may not lead to substantial wage increases if displacement effects outweigh productivity benefits. Consumer Welfare: Increases in GDP may overstate consumer welfare improvements due to investment's impact on consumption. Capital-Labor Dynamics: AI's use may increase the capital share of income, further shifting income away from labor. Investment Response: AI-related productivity gains will likely trigger only modest capital stock increases. [Caveat: above summary generated by ChatGPT, sorry!] I'm cautious -- neither too optimistic (although the engineer in me wants to believe in the AI-driven future) nor too pessimistic (but I'm also very, very realistic in broad adoption and benefits dissemination of AI). His recent chat with FT [2] is also illuminating (Sep 2024). --- [1] https://economics.mit.edu/sites/default/files/2024-04/The%20Simple%20Macroeconomics%20of%20AI.pdf https://economics.mit.edu/sites/default/files/2024-04/The%20... (Apr 2024) [2] FT audio interview - https://www.ft.com/content/e865a777-501a-4417-839e-dffd4b5206e7 https://www.ft.com/content/e865a777-501a-4417-839e-dffd4b520... (Sep 2024)