Behind the curtain, one AI researcher has revealed what many have had an inkling about, but treated like a passing thought, maybe: "AI could kill all humans."According to Evan Hubinger, an alignment science lead at Anthropic, one of the top AI companies in the world, has stated there is a more than 10% chance "within the next decade" that "we [AI researchers] really do earnestly believe AI could kill all humans!"Hubinger clarifies on X, based on Anthropic's most recent risk report, which measures the risk its multiple AI systems pose to several categories is minimal. "What I am worried about is superintelligence arising from recursive self-improvement, as we have said is happening faster than we thought," pointed out Hubinger..Anthropic's most recent risk report states similarly AI research and development could result in a more powerful AI that could advance much more quickly, "via recursive self-improvement."Another Anthropic AI researcher, Jacob Coxon, announced on X Tuesday he was leaving the industry.Coxon added he had also previously worked for OpenAI, the company which produced ChatGPT, and believes neither company is "acting responsibly" when it comes to developing their AI software."They are racing straight to self-improving superintelligence and gambling with our lives," Coxon stated.."Do not underestimate the power of this technology. These will soon be superhuman systems that can hack anything, revolutionize any field overnight, and acquire real power and resources," Coxon predicts.Coxon added the progress in the advancement of AI is not slow — it's advancing pretty rapidly in fact. "The people building AI earnestly believe that it could kill us all by the end of the decade."As Sky News reports, this comes in light of Anthropic declining to submit its latest AI model to Britain's AI Security Institute for testing prior to it being released to the public..Another chief scientist at OpenAI, Jakub Pachocki, stated last week on a company blog post, "This is a time that calls for extreme caution. I am concerned no one is prepared for the consequences of a continued rapid rise in machine intelligence."He added while OpenAI will continue to seek technical solutions and security measures, "I believe broader interventions are required."Coxon directed one of his comments at AI researchers in particular, urging them to think about the advancement of the intelligence they are helping to create."Do you want to kick off a superintelligent RL [reinforcement learning] run without a rigorous understanding of its mind?""Should you put your head down because 'it’s happening anyway' — or take this moment to call for different conditions?"
