The Anthropic researcher whose resignation made waves last month says he was 'working to automate' himself (www.businessinsider.com)

Jacob Coxon, a former capabilities researcher at Anthropic, testified before New York City lawmakers, warning that leading AI labs are now prioritizing the automation of AI research, a goal that he believes poses significant risks to humanity. Coxon revealed that during his tenure, there were set milestones aiming towards developing an automated AI researcher by 2027 or 2028, suggesting that such advancements could be imminent. He noted that as AI gains the ability to perform its own research, humans will become increasingly disengaged from the process, raising concerns about oversight and accountability in AI development. Coxon's resignation and his subsequent testimony highlighted a growing fear in the AI community regarding the pace and lack of safeguards in AI advancements. He criticized firms like Anthropic and OpenAI for their reckless approach, arguing that the startup mentality of rapid innovation could be catastrophic when applied to powerful AI technologies. His call for increased transparency and a slowdown in AI development resonated with industry leaders, including Anthropic CEO Dario Amodei and OpenAI CEO Sam Altman, who acknowledged the necessity of a more cautious approach to progress in AI capabilities.