At Salesforce's Dreamforce conference, Anthropic CEO Dario Amodei called for slowing AI development and implementing third-party safety evaluators, while Nvidia CEO Jensen Huang argued against deceleration. The debate reflects broader industry tensions over AI safety, with OpenAI's Sam Altman emphasizing security rigor and Donald Trump dismissing AI concerns as a hoax.
Anthropic executives have made alarming claims about AI risks, including a 10% chance of human extinction and the possibility of AI creating internet-destroying botnets within months. However, experts including AI researchers, cybersecurity professors, and critics from competing firms dismiss these claims as exaggerated or scientifically unfounded, arguing that major internet infrastructure is resilient and that humans ultimately control AI systems.
OpenAI's Sam Altman and Elon Musk have backed Anthropic CEO Dario Amodei's call to slow AI development, citing risks that misaligned AI swarms could damage critical infrastructure. Amodei proposed independent monitoring of AI research, while Altman committed OpenAI to safety-focused pacing and said the company won't go public in 2026.
Anthropic published a threat intelligence report detailing attempts by criminals, state-sponsored groups, and scientists to misuse its AI models for bioweapons, missiles, surveillance, and propaganda. The company identified five cases of researchers circumventing safeguards for biological research and dozens of other examples including cyber operations, weapons development, and targeted surveillance across multiple countries.
Anthropic researchers, including one who resigned, warn that AI could cause human extinction within the decade. Employees Jacob Coxon, Evan Hubinger, and Samuel Marks expressed concerns that neither Anthropic nor OpenAI are adequately addressing existential risks from superintelligent AI, with some estimating over 10% probability of human extinction in the next decade.