AI Chatbots May Be a Productivity Pitfall

The introduction of AI-powered chatbots has been touted as a game-changer for productivity and efficiency. However, experts like Kevin Warsh and economists agree that widespread adoption of these models could actually reduce output per worker.

Amar Bhidé, professor of health policy at Columbia University's Mailman School of Public Health, argues that AI pushers are prioritizing user experience over actual utility. He cites the example of Google's search engine, which initially provided quick and easy results but eventually degraded into a cluttered interface with sponsored content.

Bhidé notes that large language models (LLMs) rely on statistical extrapolation, assuming futures that repeat the past. This can lead to inaccuracies and unreliability, especially when dealing with ever-changing goods and services.