SNL Mocks AI CEO's Warning That His Own Tech Could End Humanity

Saturday Night Live opened this weekend with a sketch about the strangest thing in tech right now: the people building artificial intelligence keep warning everyone that it might be dangerous. The target was Dario Amodei, CEO of Anthropic, the company behind the Claude chatbot and one of the biggest AI players alongside OpenAI and Google. Weekend Update host Michael Che introduced him as a CEO who "stumbled through a press tour," apparently agreeing with a former employee's claim that AI could wipe out humanity.

Cast member Jane Wickline played Amodei in a big wig, stammering through answers and occasionally arguing with herself in a creepy, Gollum-like voice. "AI is the devil and I its maker," she confessed. She assured viewers that AI executives "are all on the same page here: We do not condone what we are doing." And she called AI "not a weapon, it's a tool: a tool for building weapons," before begging, "I urge you to urge me to stop."

The sharpest joke was about AI's promised benefits. Asked about curing cancer, the fake Amodei offered odds: "In 10 years, there's about a 10 percent chance that cancer won't be a problem for anyone." That number is barely better than a coin toss, and it's the kind of hedge real AI leaders use constantly.

Why should you care about a comedy sketch? Because it shows how the AI safety debate has gone mainstream. The same executives selling you AI tools are publicly unsure whether those tools are safe, which is unusual for any industry. You don't have to be a tech expert to find that odd, and now you don't have to be a comedian to joke about it either.