Physics professor warns AI self-improvement is 'worst idea in history'

fortune.com

Artificial intelligence companies are nearing "recursive self-improvement," where AI models autonomously enhance themselves, prompting a physics professor to call full autonomy "the worst idea in the history of humanity." Anthropic reported its Claude model now leads 26% of its research and development, while OpenAI aims for an automated AI researcher by March 2028. Elon Musk predicts xAI's Grok could achieve fully automated improvement by 2027. Safety concerns have split the industry, with Anthropic supporting a verifiable slowdown and OpenAI admitting it lacks knowledge on safely achieving full RSI, though it values the goal for potential safety research.


With a significance score of 6.1, this news ranks in the top 0.1% of today's 30040 analyzed articles.

Get summaries of news with significance over 5.5 (usually ~10 stories per week). Read by 10,000+ subscribers: