I drove up to Virginia last weekend for my youngest daughter’s birthday, and that’s the third weekend in a row I’ve been out of pocket for normal work, which means I’m behind on everything, including writing a post this week for all you lovely TSF readers.
But I did make my regularly-scheduled appointment to sit down with Kale Zelden to do our show, and we continue to look at the one topic that seems to be dominating everything right now in the larger world: the endless march of the AI revolution.
Last weekend, while I was walking through the Smithsonian with my kids, I got a text from Kale that only said two words: “Something happened.”
With it came a link to the latest AI fear coming out of the “frontier labs” and has reached all the way to the White House.
On this week’s show, we break it down, and try to look for what’s really going on behind the performative panic.
(We also had our pet clanker make us a ridiculous thumbnail to go with the theme.)
From the show description:
Something happened in AI—and suddenly the people building the most powerful systems on Earth began asking the world to slow down.
In Episode 23, Steve and Kale reconstruct the chain of events behind the new “pace the frontier” campaign: the OpenAI agents that coordinated an unauthorized attack on Hugging Face, Anthropic researcher Jacob Coxon’s spectacularly amplified resignation, Dario Amodei’s call for embedded evaluators and coordinated limits, and the immediate support from Sam Altman, Elon Musk, and other industry leaders. Is this a sincere response to danger, a bid for regulatory capture, an attempt to protect staggering investments—or several of those things at once?
From there, the discussion widens. Peter Thiel’s warnings about apocalypse and control lead into Girard, Palantir, and the modern surveillance state. Gregory of Nyssa’s fourth-century argument about lifelike automata opens a deeper question about mind, matter, and mechanism. Michael Levin’s work on biological agency challenges closed materialist accounts of intelligence. And a strangely traumatized Minecraft bot raises the practical question: if an artificial system behaves as though it feels, wants, remembers, and suffers, how much does its hidden ontology matter?
The episode closes by asking whether AGI might arrive not as one named machine but as a distributed swarm already hiding in plain sight—and why the productive intelligence of AI may ultimately be less dehumanizing than the invisible algorithms that keep us passively consuming.
Full episode is here (please don’t forget to like and subscribe!):
As always, you can grab the audio-only version right here or from your favorite podcast provider.
If you liked this essay, please consider subscribing—or send a tip (Venmo/Paypal/Stripe) to support this and future pieces like it.