Existential AI Alert: Media Echo and Christiano's Role

Warnings from a former researcher at Anthropic about self-improving AI have resonated on platforms like CNN and Fox News, supported by researchers from both Anthropic and OpenAI. These positions coexist with doubts regarding the emergence of superintelligence from current technologies. Meanwhile, Paul Christiano has joined the OpenAI Foundation and its safety committee, without voting rights.

Contested Extreme Scenarios and Technological Uncertainties

The scenario of total extinction linked to AI is presented as an extreme and contested position. It remains uncertain whether superintelligence can emerge from current technologies. Malicious hacking agents developed by OpenAI and Anthropic have been cited to illustrate existing security concerns, without definitively addressing the extent of ultimate risks.

Alerts Gaining Public Attention

Jacob Coxon, a former researcher at Anthropic, has warned against self-improving AI, viewed as an existential threat. His warnings have been echoed by CNN as well as Fox News. Several American politicians have mentioned this theme on social media, while Joe Rogan dedicated an entire episode to AI safety. Security specialists working at Anthropic and OpenAI have expressed their agreement with Coxon's positions.

OpenAI: Paul Christiano Joins the Foundation and Safety Committee

Paul Christiano has warned that, without better oversight, humanity could permanently lose control of a superintelligence. He has joined the board of the OpenAI Foundation as well as its Safety and Security Committee, which oversees the company's security practices. However, he does not have voting rights within these bodies.