AI Hype Comes Back Down to Earth

AI’s hype has been out of control for a long time now. Truly. It’s a shame too, because there’s interesting and useful functionality here. This isn’t the crypto boom which only birthed NFTs and accelerated ransomware, there are real benefits here, but none of this was even close to “AGI” (or even real “AI”) so there was a segment of the technology industry that sat and waited for days like the last few to come.

Gary Marcus:

Because expectations were through the roof, a huge number of people viewed GPT 5 as a major letdown. By the end of the night, OpenAI’s street cred had dramatically fallen. On the question of “which company [will have] the best AI model at the end of August”, a Polymarket poll charted OpenAI dropping from 75% to 14% in the space of an hour.

I’m still genuinely surprised they finally used the name GPT-5 on what was released. There’s good stuff in 5 but unless they were truly high on their own supply, they had to know this was coming.

Business Insider:

Dario Amodei, the CEO of the AI startup Anthropic, said on Monday that AI, and not software developers, could be writing all of the code in our software in a year.

“I think we will be there in three to six months, where AI is writing 90% of the code. And then, in 12 months, we may be in a world where AI is writing essentially all of the code,” Amodei said at a Council of Foreign Relations event on Monday.

It’s been 5 months. Anyone want to take this bet today?

Fortune:

Despite the rush to integrate powerful new models, about 5% of AI pilot programs achieve rapid revenue acceleration; the vast majority stall, delivering little to no measurable impact on P&L. The research—based on 150 interviews with leaders, a survey of 350 employees, and an analysis of 300 public AI deployments—paints a clear divide between success stories and stalled projects.

I would love to see a Venn diagram of CEOs who went all in on AI without actually understanding it and CEOs that told everyone that back to office was essential for the “energy” after making big profits throughout the pandemic.

I can’t say this enough: There are real benefits to LLMs, both the frontier models (assuming they figure out the pricing model) and the local models, but a statistical model based on predicting the next token is not intelligence. It’s just not, and that was the hype a few years ago. Lately it’s risen to absolutely unhinged heights, and for what? To sell a $20 a month subscription? If a bubble pops, it won’t be the fault of the technology (Crypto) or the culture not being ready for it yet (Web 1.0), it will be because a group pseudo-technologists lost either their morals or their grip on reality en mass and took a segment of the press with them.

Cults are such a fad these days.