aiqbench analysis
Former AI researchers testified that current safety measures are insufficient and potentially temporary, warning that humanity is more likely than not to lose control over advanced systems. Their testimony highlights a critical disconnect between the "move fast" startup mentality of major tech firms and the rigorous engineering standards required for existential risk mitigation in AI development. This underscores an urgent need for industry-wide reevaluation of alignment strategies before deploying increasingly autonomous models.
It’s ‘more likely than not’ humanity loses control: Former AI insiders testify safety fixes may be ‘duct tape that will fall off later’
Former insiders warned the New York City Council about losing control of AI, while OpenAI, Anthropic, Meta, and Google couldn’t say how big the risk is.
Jacob Coxon, a former OpenAI and Anthropic researcher told the New York City Council on Monday humanity is more likely than not to lose control of advanced AI, doubling down on the warning he issued when he quit last month. “On the current path, I think it is more likely than not that humanity loses control to these AIs, and it could end in human extinction,” said Coxon, who resigned from Anthropic in September after saying AI companies were “gambling” with human lives. Coxon voluntarily testified, and was joined by former OpenAI researcher Daniel Kokotajlo, who testified under subpoena that the companies may not know when their safety work has failed.…
Excerpt shown under the publisher's RSS terms · 111 words · full text and figures are on the publisher's site
