- Jacob Coxon announced his resignation from Anthropic, accusing both Anthropic and OpenAI of "gambling with our lives" and warning that AI could "kill us all by the end of the decade."
- Evan Hubinger, Anthropic's alignment lead, responded to Coxon's resignation by stating he expects a more than 10% chance of extinction due to AI risks.
- Julie Steele, a member of OpenAI's safety team, echoed concerns on X, stating, "we need to slow down" in response to Coxon's warnings.
- Samuel Marks, an Anthropic researcher, also expressed concerns on X, indicating that AI developers believe their technology could cause human extinction in the near future.
- Jasmine Wang, an OpenAI alignment researcher, warned about the dangers of recursive self-improvement (RSI), stating there is no viable scientific plan to address these risks.
- Jakub Pachocki, OpenAI's chief scientist, called for extreme caution in a blog post, expressing concern that no one is prepared for the consequences of rapid AI development.
- Paul Christiano, former head of CAISI, expressed a risk of catastrophic loss of control due to rapid AI capabilities development.
- Rep. Lori Trahan urged Congress to act, stating that it is past time for lawmakers to address AI safety concerns.
- The resignation of Jacob Coxon went viral, transforming initial fears about AI into a widespread concern.
Jacob Coxon, an Anthropic researcher, has resigned, citing concerns that AI labs like Anthropic and OpenAI are "gambling with our lives." He warned that AI could potentially lead to human extinction by the end of the decade.14
In response, Evan Hubinger, Anthropic's alignment lead, acknowledged a greater than 10% chance of such outcomes occurring. Other researchers, including Julie Steele from OpenAI, have echoed the need for a slowdown in AI development, emphasizing the risks involved.23
Samuel Marks, another Anthropic researcher, stated that many AI developers share concerns about the technology's potential to cause catastrophic outcomes. He noted that the more senior the employee, the greater the concern.
Jasmine Wang, an OpenAI researcher, highlighted the dangers of rapidly advancing AI, stating, "It's hard to overstate how dangerous speeding towards RSI is."5
OpenAI's chief scientist, Jakub Pachocki, expressed that the current pace of AI development could lead to significant capability jumps, urging for extreme caution. He stated, "I am concerned no one is prepared for the consequences of a continued rapid rise in machine intelligence."6
Former safety head at the U.S. Commerce Department, Paul Christiano, warned of a meaningful risk of catastrophic loss of control due to rapid AI advancements.7
The discourse surrounding AI risks has intensified, with many researchers advocating for a more cautious approach to development, reflecting a growing consensus on the potential dangers posed by unchecked AI progress.
“Evan Hubinger, Anthropic's alignment lead, puts the chance of AI extinction at over 10%, while OpenAI's Julie Steele says 'we need to slow down.' Rep. Lori Trahan urges Congress to act, noting 'safety researchers are resigning' and 'companies are racing ahead anyway.'”
CNBC
Interconnects AI
