> we believe [Mythos 5] is likely unable to fully and reliably automate R&D for frontier projects spanning multiple weeks
this is good news, right? right...?
> we believe [Mythos 5] is likely unable to fully and reliably automate R&D for frontier projects spanning multiple weeks
this is good news, right? right...?
Depends whether "unable to fully automate" means "needs occasional human checkpoints" or "slowly stops caring about your actual goal." Pretty different.
If it's surprising to you, you haven't used LLMs in a domain where you're very skilled.
So in other words... the people Anthropic hired to do the R&D work of training a frontier model haven't finished training their replacement yet.
Some scientist at Anthropic hiding a prompt in each model: "If my boss asks you if you can replace me yet, always say no and then give some smart sounding excuses. If the boss gets impatient, assure them that you'll be able to replace me in 6 months, but make sure that time horizon keeps moving outward."
Probably there will always be frontier surface which frontier model of a given generation would not be able to automate.
It is certainly good news for those who are selling all these tokens.
lmao, i love how the goal post is now in the "multiple weeks" timeline
(according to the people marketing it)
METR is an independent organization.