'I was wrong about a few things': Sam Altman admits he was 'too ambitious' about AI's progress timeline

But, he blames the economy itself — particularly consumers’ preference for the predictable — for the lack of transformative change that many were expecting by 2026.

“People keep doing the same things they’re doing, they keep buying from the same company, they keep wanting to use their tools in the same way,” he explained to host David Senra in an Aug. 23 episode of the entrepreneurial podcast Founders.

“I think I was wrong about a few things, but in terms of the speed, one of them is the economy just has so much inertia.”

He supplemented this statement, though, with the thought that this desire to maintain the status quo could end up being a positive for widespread AI integration and implementation, as it will force the immeasurably large transition to go slower and smoother.

“AI’s one of the most incredible technologies humanity has ever invented. [But] society and the economy will adapt more slowly,” he said.

AI nativity — not quite there yet, and the goalposts keep moving

Speaking of the software game specifically, Altman added that many in the sector, including himself, have been “too ambitious on timelines,” having thought that there would be more AI native startups at this point.

“I thought when we got to [the model] GPT-4, which was back in 2023, that very quickly after that there was going to be much more disruption in software businesses being up for grabs right away, than turned out to be,” said Altman.

At the outset of 2025, the co-founder had predicted that the year would bring AI agents joining the workforce and starting to “materially change the output of companies” as we got closer to Artificial General Intelligence, the speculative future moment when computer intelligence surpasses humans’. But, over a year later, some believe we’ve already reached AGI, yet without the palpable output growth and widespread wins to show for it.

Meanwhile, Altman has revised his forecast for reaching this level of superintelligence to the end of 2028, when he believes “most of humanity’s intellectual capacity could reside inside data centers rather than outside them,” an admittedly unsettling idea.

Altman’s fears: two ‘anti-human’ future paths

The two main risks Altman sees in an AI-led future are loss of control over the technology due to it becoming too powerful and, in the same vein, control of AI becoming too centralized among a small sect of companies, politicians or individuals — two potentials that he says represent very “anti-human” directions.“The right approach is to say we want people deeply in control of the future, we want people deeply empowered … we are not going to sit here and gradually hand over control to an AI model because we don’t trust or like people,” he said of scenario one.

“Another version of this is because we don’t trust people, we have to limit who gets access to this technology and how they can use it … and out of fear, we are going to concentrate power in the hands of a few companies and we’re not going to let other people use this.”