korrents

Our readingWhat is a korrent?

If AIs can now build and improve game engines from their own rollouts, it is reasonable to assume they have gotten much better at building the RL environments used in their training.

Drawn from what Dwarkesh Patel said

What this subject means

reinforcement learning Training by reward: environments, verifiable tasks, value functions, and whether it works or merely beats what came before.

What Dwarkesh Patel actually said

  1. Dwarkesh Patel Interviewer; the Dwarkesh Podcast runs long
    If AI's can now do the work of building these game engines, including even the work of improving these engines based on their own experience navigating through these virtual worlds through hundreds of parallel rollouts, it's reasonable to assume that AIs have gotten much better at building and improving the RL environments that go into their training.

    Watch from 14:11 plays here

    ↗Which AI is the best liar?youtube.com

    5 Nov 2026 · video · 18m · spoken · machine transcript

    All 8 korrents from this recordingthis one is 7th

Added to korrents 8 Nov 2026 · How quotes work · Something wrong? Tell us

Do you hold this korrent?Do you also believe this?

I hold this too

Related korrents

Shared subject first, then similar wording. They may agree or disagree.

See this claim on the map