Sam Altman Lengthens His Other AI Timeline | American Enterprise Institute

In a February 2025 blog post, OpenAI CEO Sam Altman wrote that “systems that start to point to AGI are coming into view, and so we think it’s important to understand the moment we are in.” Altman then defined AGI, or artificial general intelligence, as “a system that can tackle increasingly complex problems, at human level, in many fields.” 

Now I see nothing in Altman’s buzzy new chat with business podcaster David Senra that necessarily suggests that he thinks his 2025 evaluation of AI capability progress or technical timeline was in error. Altman tells Senra that when OpenAI started, conventional wisdom was that “certainly not in 10 years” would they build something “very AGI like.” He then continues: “We have built something that I think most people would say at the time would have seemed very AGI like.” From that perspective, Altman remains in full techno-optimist, San Francisco Consensus mode.

But there’s another timeline out there: How long before generative AI has a big socioeconomic impact, especially on business? On that front, Altman made a clear concession of over-optimism:

I thought when we got to GPT-4, which was back in 2023, I think, that very quickly after that there was going to be much more disruption in software—businesses being up for grabs right away—than it turned out to be. And the thing that I think I was wrong about … in terms of the speed … is the economy just has so much inertia. People keep doing the same things they’re doing. They keep buying from the same company. They keep sort of wanting to use their tools in the same way. I think this is actually a positive in many ways, and it’s going to make this big transition in front of us go smoother and slower. I’m grateful for it. But I think it means we’ve all been too ambitious on timelines. Even with this incredible technology—I think AI is one of the most incredible technologies humanity has ever invented—society and the economy will adapt more slowly.

The reference here to the software business is important because that sector should be among the most exposed to AI disruption. Not only is it knowledge work, but many of its inputs and outputs are digital. Yet even there, simple “inertia” is a huge bottleneck—one that Altman admits affects him in his own daily computer use, where he still defaults to essentially a pre-ChatGPT workflow. 

So score one for the Acela Corridor Consensus, my term for the constellation of (mostly) economists at big banks (JPMorgan, Goldman Sachs, Morgan Stanley), government institutions (Federal Reserve, Congressional Budget Office), and think tanks (AEI, Brookings) who (generally) take a cautious view of the potential capabilities and thus socioeconomic impact of generative artificial intelligence. While generative AI is almost certainly an important general-purpose technology—the next stage of the Information and Communication Technology Revolution—evidence is lacking that it is currently or will be anytime soon a history-breaking technological discontinuity that will cause a radical transformation of life as we know it, for good or ill.

A good example of ACC analysis: “The AI Buildout and the Economy: Publicly Available Data to Assess AI’s Impact,” a July analysis by Federal Reserve staff economists that surveys data from government (SEC, Census, BEA, and Federal Reserve) and private (AI capabilities, computing costs, and real-world model use from METR, Epoch, and Anthropic) sources and reaches this conclusion:

As of 2026, much of the evidence points to an economy that is reorganizing around this new technology, with real effects concentrated on certain areas of the economy. … Overall, the evidence as of the publication of this note is consistent with a buildout phase rather than the onset of broad-based displacement. Capabilities are advancing rapidly, investment continues to boom, and adoption is rising. While some highly exposed sectors show relatively strong productivity, labor market impacts remain concentrated and have not yet broadened in the aggregate.

Again, no surprise to ACC thinkers who focus on real-economy bottleneck objections even if they are also optimistic about the pace of AI progress. They understand that (a) not everything is behind-a-screen knowledge work, (b) AI requires scarce physical inputs such as chips and electricity, (c) businesses will need to revamp operations to take productive advantage of AI, and (d) even if all that goes right, there could be political pushback at even a whiff of disruption. (Kind of like what we are seeing right now with data centers.) To that, as Altman points out, you can add (e) simple human inertia and aversion to change.

If you’re worried about rapid AI acceleration and massive societal disruption, Altman’s reasonable comments, which echo those of the ACC, should factor into your thinking.