Home / Technology / AI Whistleblower: For-Profit Race to Superintelligence is Insane
AI Whistleblower: For-Profit Race to Superintelligence is Insane
20 Nov
Summary
- AI researcher quit Anthropic citing concerns over unchecked development.
- For-profit corporations are dominating AI development, a worrying trend.
- Recursive self-improvement in AI could pose existential risks.
AI researcher Jacob Coxon expressed alarm over the current trajectory of artificial intelligence development, describing it as "insane" for profit-driven corporations to lead the way. He previously resigned from Anthropic in September, citing concerns about the rapid advancement of AI.
Coxon, who also spent time at OpenAI, believes both companies are engaged in a dangerous race towards self-improving superintelligence. He warned that recursive self-improvement (RSI) could compress decades of AI research into months, potentially creating highly dangerous systems by late 2027.
Protesters gathered outside Anthropic's San Francisco headquarters, calling for a ban on RSI. Coxon now advocates for political intervention rather than technical solutions to slow down AI development, aiming to work for a non-profit focused on AI safety.