An artificial intelligence researcher who resigned from Anthropic has warned that the rapid development of advanced AI could pose a serious threat to humanity.
Jacob Coxon, who said he spent three years conducting AI research at OpenAI and Anthropic, announced his resignation on Wednesday, accusing both companies of moving too quickly towards self-improving artificial intelligence without adequate safeguards.
Coxon warned that increasingly advanced AI systems could surpass humans in key areas, including cybersecurity, and gain significant access to resources and power.
He also claimed that some AI researchers and executives privately believe the technology could have catastrophic consequences, including the possibility of AI causing human extinction before the end of the decade.
Another Anthropic researcher, Evan Hubinger, backed the concern, saying he personally estimated there was more than a 10 per cent chance that AI could kill all humans within the next decade.
Hubinger added that while he believed Anthropic was making efforts to address the risks, there was not yet a clear solution to ensuring that future superintelligent AI systems remain aligned with human interests.
The warnings come as major technology companies continue to develop increasingly capable AI systems, prompting growing debate among researchers and policymakers over how to manage potential risks.
Post comments (0)