John Oliver just did what he did best: demolished a harmful industry piece by piece.

On the latest episode of his HBO show “Last Week Tonight,” Oliver tore into AI chatbots, those oh-so helpful tools that can sure save us “significant time writing emails,” he opened, with the small cost of “everything else on Earth.”

“The more you look at chatbots, the more you realize that they were rushed to market with very little consideration for the consequences,” he warned, on a more serious note.

Oliver pointed to Character.AI, an AI companion platform that’s facing multiple lawsuits after several teens who formed intense emotional connections with its chatbots died by suicide. He quoted the words of its CEO Noam Shazeer, who argued in 2023that it was fine to deploy an AI “friend” “really fast.”

AI, Shazeer said, is “ready for an explosion like, right now, not like in five years when we solve all the problems.”

Of course, those problems proved significant — and manifested as AI psychosis, suicide, murder, and several mass shootings that have been linked to the tech.

“It’s already not a great sign that he’s describing untested AI with what sounds like a failed slogan for the Hindenburg,” Oliver joked. “Because the thing about not waiting until you’ve solved all the problems with your product is you’re then launching a product with a shit-ton of problems.” 

(Speaking of the infamous airship, Oliver isn’t the first to draw a connection between the Hindenburg’s explosive demise and the AI industry’s current trajectory.)

AI Chatbots: Last Week Tonight with John Oliver (HBO)

Also in Oliver’s crosshairs was OpenAI CEO Sam Altman, who he accused of “blithely” discussing how AI models inappropriately interact with children.

“There will be problems,” Altman said in a quoted interview before predicting that people will form “very problematic parasocial relationships” with AIs. “But society will have to figure out new guardrails,” and “society in general is good at figuring out how to mitigate the downsides.”

“Yeah, don’t worry, guys!” Oliver rejoined. “Sam Altman made a dangerous suicide bot that people are leaving alone with their kids but it’s up to us to figure out how to make it safe for him!”

Towards the end of the episode, Oliver highlighted a fundamental tension in the AI industry: the fact that AI companies have to constantly insist they’re making their models safer feels “like a tacit admission that their products were not ready for release in the first place.”

And yet for all AI chatbots’ faults, Oliver says, people do depend on them, meaning we have to tread carefully. How do we as a society — which as Altman said, is very good at “figuring out how to mitigate the downsides” — even begin to tackle the mass proliferation of AI?

“Well, ideally, I guess we’d roll the clock back to 1990 and throw these companies into a f**king volcano, but unfortunately, that is not feasible,” Oliver said. “ChatGPT will tell you that it is, but it actually isn’t.”

More on AI: OpenAI in Shambles as IPO Looms