Comment by agazi (mikael-1) on Would catching your AIs trying to escape convince AI developers to slow down or undeploy? · 2024-09-06T19:58:23.678Z · LW · GW

It's very fascinating to consider how the costs of undeploying would be analyzed in the heat of the moment. If we consider the current rate of LLM adoption in all parts of the economy over the next few years, one could foresee a lot of pipelines breaking if all GPT6 level models get removed from the api.

Definitely not a new comparison but this scenario seems similar to the decision to shut down the economy at the onset of Covid.

Comment by agazi (mikael-1) on Instruction-following AGI is easier and more likely than value aligned AGI · 2024-05-18T16:53:56.570Z · LW · GW

I think we can already see the early innings of this with large API providers figuring out how to calibrate post-training techniques (RHLF, constitutional AI) between economic usefulness and the "mean" of western morals. Tough to go against economic incentives

Comment by agazi (mikael-1) on Instruction-following AGI is easier and more likely than value aligned AGI · 2024-05-18T16:53:44.045Z · LW · GW