Press Space to continue
Strong mechanistic interpretability unlikely within one year
Press Space to continue
I don't think there will be strong+working mech interp in <1 year. Model internals methods could pareto dominate cot monitors in a year via cot losing ~all value, but this isn't exactly encouraging. Depending on AIs decoding other AI's opaque activations for oversight is spooky
tszzl@tszzlSep 2
*monitor-ability* is the invariant that must be preserved, and the real solution will be via strong mechanistic interpretability. i predict in the next year there’ll be mechinterp monitoring that’s pareto optimal to cot monitors