The speaker expects many bugs to be demonstrable convincingly at smaller scale, making those demonstrations usable for training AI systems.

“like a moderate scale compute experiment where you like spin up the whole distributed infrastructure and then run it but oftentimes i think you'll be able to demonstrate it pretty convincingly um at smaller scale in a way which you could actually train on um and so my sense is that like it will not necessarily like i think it wouldn't be very surprising if right now people have rl environments where they like you know introduce a subtle bug into some training recipe train the ai to point out the subtle bug and then have like you know a rubric where they're like did it actually find the right bug”