Computer Scientist Says AI May Never Achieve Human-Level Intelligence

A leading computer scientist argues that artificial intelligence may never achieve human-level intelligence because it fundamentally cannot acquire tacit knowledge—the unspoken, intuitive understanding that underpins common sense, culture, and practical expertise. The article explains that while AI excels at processing explicit data and pattern recognition, it lacks the embodied experience and social context necessary for true general intelligence. This perspective challenges the prevailing optimism in the AI industry about reaching artificial general intelligence (AGI) in the near future. The scientist points to limitations in current deep learning models, which rely on vast datasets but cannot replicate the nuanced learning that humans gain through lived experience. The piece does not name a specific researcher or institution, but it reflects ongoing debates in the field about the path to AGI. The argument has implications for AI research funding, corporate strategy, and public expectations about AI capabilities.

Global Impact

This story has moderate technological and economic implications. If the thesis gains traction, it could temper the current AI investment frenzy, redirecting billions in R&D from AGI pursuit to narrower, more practical AI tools.