descriptive claim
Amortizing the amplified model's behavior into a probe produces a very cheap, fast-running component that can serve as the core of a reinforcement-learning loop.
Tom McGrath · Designing How AI Grows — Tom McGrath · 2 Sept 2026
In their words
The transcript segments this claim was taken from
“And now we have something that's extremely cheap to run.”
“and faster run and so it can be like the core of an RL loop”Play from 42:23Report a misattribution
90%how strongly it was put, not whether it is true
82%importance to the conversation
96%confidence the claim was read correctly