Posts

How are Those AI Participants Doing Anyway? 2025-01-24T22:37:47.999Z

Comments

Comment by mushroomsoup on Instrumental Goals Are A Different And Friendlier Kind Of Thing Than Terminal Goals · 2025-01-26T19:30:52.214Z · LW · GW

This comment raises some good points, but even "there will be a natural pressure for [subprocesses] to resemble a corrigible agent" seem to be debatable. Again consider the restaurant setting. It is sometime necessary for restaurants close temporarily for renovation to increase the seating capacity, upgrade equipment, etc." The head chef who decided to renovate will be making the instrumental goals of all the other chefs (make a good food, earn money to stay alive) untenable while they are furloughed. More generally, progress towards terminal goals is not monotonic and thus only focusing on the local topology of the optimization landscape might be insufficient to predict long-horizon trends. 

Comment by mushroomsoup on Orienting to 3 year AGI timelines · 2024-12-27T01:55:48.648Z · LW · GW

Thanks! The linked article is exactly what I was looking for. Assuming "nationalization" means something like "soft nationalization" does make the timeline seem a lot more plausible. 

Comment by mushroomsoup on Orienting to 3 year AGI timelines · 2024-12-23T00:27:11.411Z · LW · GW

According to the timeline of the post, AGI will take place during the Trump presidency and much of the nationalization efforts will need to be lead by his administration. However, that seems antithetical to the administrations general ethos of deregulation (at least in the banking and energy sectors). Would it be possible to explain which avenues would lead Trump to nationalize, for example, OpenAI or Antropic?