Politics Gets Interested In Those Trying Not To Die

The newsletter "Politics Gets Interested In Those Trying Not To Die" recounts a month in which, it says, the world took notice that AI might kill everyone. It cites Jacob Coxon's resignation as setting off a preference cascade, and states that Anthropic CEO Dario Amodei wrote that the frontier must be paced, with Sam Altman, Elon Musk and Demis Hassabis agreeing.

According to the account, the shift reversed after Trump "went full 'hoax' on existential risk," which the newsletter attributes largely to Jensen Huang, Mark Zuckerberg and David Sacks instilling paranoia and fears of economic problems. The newsletter says Trump conflated existential risk with attacks on data centers and has since doubled down, with others in the White House echoing talking points.

On September 19, around 10:12am, Trump published a statement listing past "Hoaxes" and adding "the decimation, or destruction, of AI." He said the "crazed Data Center attack has largely failed" and that he would not hinder the industry's growth, adding that bad actors would be addressed through the existing criminal and civil justice system. He announced he was forming the "AI Force," much like Space Force, and would soon name an AI "Czar."

Trump also floated renaming AI, polling readers on "Superior Intelligence (SI)," "Extreme Intelligence (EI)" or "Supreme Intelligence (SI)." He later removed "Supreme Intelligence," saying it was losing badly to the other two, and restarted the vote.

Citing a New York Times report by David E. Sanger, Jonathan Swan, Cecilia Kang and Dustin Volz, the newsletter says people who discussed the subject with Trump in private say his outbursts are rooted in worries that a slowdown in the AI boom fueling the economy could cause a market crash and recession. It also cites concerns among Treasury Secretary Bessent and chief of staff Susie Wiles about AI disruption, and Wiles organizing an informal group on risks from more powerful models.

The newsletter quotes JD Vance saying that if you are building "Frankenstein," you should stop rather than ask for regulation, and Michael Kratsios saying companies that believe a technology is unsafe can simply stop. It also quotes David Sacks claiming a million jobs created around the AI buildout, a figure the newsletter disputes.