Yesterday our director, Robert J. Marks, offered a short, real-world history of impressive progress in AI that — of course — falls far short of the superintelligence hype. Last week, technology consultant Jeffrey Funk also pointed out here that 95% of AI initiatives fail to achieve a positive return on investment.

Meanwhiler, at Fast Company, Grace Snelling offers a look at Brian Patrick’s “AI Executive Insanity Series,” which methodically tracks nonsense talked by AI execs and posts it on social media:

There’s the clip of Oracle founder Larry Ellison admitting to building body cameras that record users while they’re in the bathroom. A sound bite of SpaceX CEO Elon Musk advocating for merging humans with AI. Even an interview with Friend CEO Avi Schiffmann, in which he compares his relationship with AI wearables to peoples’ relationships with God.

“Meet the creator tracking outlandish claims from AI executives every day,” July 1, 2026

So far, Patrick’s group Panodime has 181 of them:

“I think just more and more the way I heard them speak, it was so disturbing to me,” Patrick says. “The more I listened to it, the more I realized that I didn’t know the depth of it—and I felt like most people weren’t aware of the depth of it. Most people just aren’t aware of how many standard deviations these people are away from the mean, and how their views of the world are so warped in very disturbing ways.” Every day”

Ultimately, he wants more democratic control of AI though just what that might mean is unclear. Most of the AI apocalypses aren’t going to happen anyway because failure is the price people pay for divorcing reality.