< Back to all clusters
[TECHNOLOGY] · United States · 4 sources

started · updated

Jacob Coxon warns of AI risks and advocates for ‘kill switch’

Former AI researcher Jacob Coxon has issued warnings regarding the rapid development of superintelligent artificial intelligence, citing existential risks to humanity. Coxon, who previously worked at both OpenAI and Anthropic, has advocated for the implementation of a ‘kill switch’ for AI systems, though he noted such a measure might become ineffective if an AI performs an internet-wide hacking run.

Coxon suggests that while AI labs may be genuine in their desire to slow development, they are currently trapped in a competitive race. He has called for increased self-regulation within the industry to manage these risks.

Adding to these concerns, Evan Hubinger, Anthropic’s head of Alignment Science, indicated there is a 10% probability of a catastrophe for humanity within the next decade due to the technology currently under development.

Entities

Anthropic · Dario Amodei · Evan Hubinger · Jacob Coxon · OpenAI

Claims

What the coverage asserts, and how many sources carry each claim.