This may be the month the AI spell finally broke.

For more than two years we’ve been told that the AI takeover was unstoppable. Inevitable. Talking with tech hypesters felt like confronting the hivemind of the Borg: Resistance is futile.

Over the past four weeks, though, the weather has changed. The public taste for AI, excited by those first encounters with ChatGPT and other magical talking bots, is beginning to sour.

Consider:

This is not the end of AI. This is the end of the starry-eyed glorious-digital-future phase of artificial intelligence. This is the beginning of the end of the hype.

It was easy to believe AI could do anything after a first encounter with ChatGPT. The release of Sora sent the entire film industry into a panic.

I’ve seen the tide beginning to turn. Two years ago most state legislators scoffed at the idea of raising guardrails around AI. “This is great stuff!” they’d say. “ChatGPT answers all my questions. What’s the worry?”

You don’t hear that so much anymore. Lawmakers have read the accounts of teens killing themselves with the encouragement of chatbots. They’ve seen the detrimental effects on their own kids. They hear the outrage about data centers from their hometown constituents.

They’ve also witnessed the massive overreach of the tech industry. Over the past year the companies behind the AI rollout tossed their ethics to the wind, partnering with Trump in a bid for unlimited power and profit. Their repeated attempts to pass a law exempting them from all laws that might impose the slightest safety standards on this powerful new technology showed them for what they are: Unbound by any moral sense whatsoever.

The AI industry isn’t about to collapse. But the go-go phase of hysteric investment and unlimited spending may have peaked. We saw this in earlier cycles of tech exuberance: At a certain point a startup’s massive burn rate becomes unsustainable. By some estimates it cost OpenAI $15 million a day to create Sora’s funny five-second video clips. Your $20 monthly subscription ain’t covering that cost.

Both OpenAI and Anthropic are hoping to launch IPOs later this year. That will bring a lot more hard-eyed scrutiny to their products, their costs, and the real (or phantom) value AI is returning to the businesses adopting their products.

Some tech CEOs can see a reckoning on the horizon. In late January, Microsoft CEO Satya Nadella warned that AI will start losing public support unless it starts to “do something useful that changes the outcomes of people and communities.”

“We will quickly lose even the social permission to take something like energy, which is a scarce resource, and use it to generate these tokens [that power AI],” he said.

In other words: We’d better find beneficial uses for AI before the public turns on us.

Which was not the worst warning the CEO of Microsoft could send to his industry.

Of course, Nadella immediately undermined his own message. Instead of pushing tech companies to actually make AI useful, he proposed pushing more hype and fear.

"The demand side of this is a little bit like, every firm has to start by using it," he said. "People need to say, 'Oh, I pick up this AI skill, and now I'm a better provider of some product or service in the real economy.’"

In other words, Nadella urged the tech industry to encourage job seekers to learn AI skills in order to become more employable. Even as the industry continues to tout AI as a way to shed labor costs and, you know, the people who are employed to carry out the labor.

If you enjoyed last week’s post about Shy Girl and book publishing in the age of AI, check out Jonathan Evison’s podcast, A Fresh Face in Hell. Jonathan had me on last week to talk more about books, AI, authorship, and the value of authentic human creation.