The AI Slowdown Debate | Diet TBPN
John Coogan and Jordi Hays join TBPN to break down Dario Amodei's essay proposing ways to slow AI development and the legal, political, and competitive problems that follow.
The gist
-
Coordinating AI labs on safety standards would likely violate the Sherman Antitrust Act, making any formal pacing arrangement legally difficult to pull off.
-
Third-party evaluators like MITER may lack real independence because their staff often come from the same labs they are supposed to oversee.
-
Heavy regulation aimed at frontier labs could hit small startups hardest, forcing tiny teams to spend years on compliance instead of building products.
-
Trump has publicly called AI safety concerns a conspiracy and is pushing for acceleration, which limits how much formal oversight the White House will support.
Also mentioned
Related episodes
Episodes covering similar topics