I have recommended David Shapiro to a bunch of people. I have learned a lot from his content, and I have used some of his materials for the Towards Chronic Health workshop and on my posts here and on Your Personal Singularity.

But I’m afraid I have to push back on his more recent output somewhat vehemently.

Overall, I have seen Dave is leaning heavily into the post-labor economics, an area of his ongoing research, or rather speculation. I think he may have a book coming out on that topic, too. That’s all fine.

Where he’s lost me now is the uber-naive acceptance of PR spin from OpenAI in the context of the IMO, which was picked up by the press, plus and all influencers.

I mean, Dave said he actually believes that the OpenAI team basically stumbled into these results by pure chance. Sure 🙄.

Then OpenAI’s PR team also didn’t hear about the press embargo and released their results earlier than everyone else, making it look like they got their first. Pure coincidence, I guess.

IMO organizers had requested participants—particularly AI developers—to withhold the publication of results until after the official closing ceremony and embargo lift date (around July 28, 2025) to allow human competitors their deserved recognition. OpenAI released its results on July 19, several days before the embargo lifted, which an IMO coordinator described as “rude and inappropriate” because OpenAI was not formally cooperating under the IMO’s testing agreement.

If you care to learn a bit more, read Georg’s post here, which is a more ruthless assessment of “best practices” of benchmark hacking and the continued amplification of this PR spin by volunteering influencers.

I would still stand by the value and usefulness of the materials from Dave that I recommended, but use your own judgment when it comes to “AI is going to solve and automate everything.” and “AGI around the corner.” kind of statements.