AI Generated Image
‘Gambling with our lives’: AI researcher quits Anthropic with dire warning about safety
politico.eu • • world
Key Points:
- Jacob Coxon, an AI researcher formerly at Anthropic and OpenAI, resigned citing concerns that these companies are "gambling with our lives" by advancing AI technology too rapidly.
- Coxon warned that AI systems are approaching superhuman capabilities, able to hack anything and revolutionize fields overnight, with progress showing no signs of slowing.
- He criticized Anthropic and OpenAI for racing toward self-improving superintelligence, a feedback loop where AI models create increasingly powerful successors, potentially surpassing human intelligence.
- Evan Hubinger, Anthropic’s lead on AI alignment, supported Coxon's concerns, estimating over a 10% chance that AI could kill all humans within a decade and highlighting the lack of plans to ensure AI remains aligned with human values.
- In response to these risks, U.S. Senator Bernie Sanders announced plans to introduce legislation aimed at banning the development of superintelligent AI systems.
