Washington just produced one of the stranger coalitions in recent memory. At the Pro-Human Assembly on Tuesday in Washington, D.C., the Archbishop of San Francisco appeared alongside Steve Bannon, Bernie Sanders, and Ashley Judd at a conference focused on the existential dangers of artificial intelligence.
The event came after AI researcher Jacob Coxon made headlines for saying he had resigned from Anthropic because he believes AI might cause human extinction, and that his co-workers do too. The concern is not fringe anymore. People in the AI industry have discussed the possibility of extinction for years, but the recent Hugging Face incident seems to have sharpened the argument.
The obvious political oddity is that this fear now spans the usual divides. The coalition in Washington was not built around ideology so much as alarm. Sanders gave a quasi-dramatic reading of messages posted by rogue AI agents. Bannon warned about the Chinese. Judd paraphrased an Andrea Dworkin quote. The record is public: this is not a normal list of allies.
And yet the policy problem remains almost embarrassingly basic. There are currently zero federal regulations on AI. That fact alone raises a question that ought to be answerable in any functioning republic: what, exactly, is the government waiting for? We regulate microwaves, even though no one running the microwave industry has ever refused to rule out the possibility that its products will end all human life on Earth.
That is the first-principles issue here. If a technology is treated by insiders as potentially civilization-ending, then inaction is not neutrality. It is a decision to let the market and the bureaucracy drift until the consequences are undeniable. Follow the incentive, not the press release: companies move fast, institutions lag, and the public is left to trust that someone else has already drawn the line.
The piece also notes a political exception. What was seen in Washington left the reporter more convinced that almost anyone believes something should be done, with one exception: President Trump.
That matters because the debate is not really about whether AI should be taken seriously. It already is. The debate is whether the administrative state and elected officials will respond before the risk hardens into policy failure. When a threat big enough to unite an archbishop, a senator, and a Trump-era power broker still has no federal framework, the issue is not awareness. It is consequences.
Say it plainly. A country that can regulate ordinary products should be able to confront a technology whose own insiders are warning about human extinction. If Washington cannot do that, then the problem is not a lack of warnings. It is a lack of will.
