"You could have a situation where the model understands what chain of thought is and that people are observing it. This is all in the pre-training data."
“One of the major takeaways from the incident is that people underestimated the AI.
And we never want to be in a situation again where we underestimate the AI.”
"Things like chain of thought monitoring buy us time, and they can tell us if we're on the right path. But at the end of the day, we really do need to solve the alignment problem."
@polynoamial
Historian Jared Diamond: "If Genghis Khan had been born 10 years later or 200 miles further away, we wouldn't know anything about him."
Jared Diamond: "Genghis Khan created the largest land empire in world history, all the way from China to Poland. And yet, there were other people who tried—Attila the Hun and Tamerlane the Turk. Why did Genghis Khan succeed?
Well, within the last 10 years, students of tree rings have figured out that on the Mongolian steppe, the wettest time in the last 2,000 years was the decade of Genghis Khan's birth. If he had been born 10 years later or 200 miles further away, we wouldn't know anything about Genghis Khan. But he happened to have been born at the right time and place because with all that rain, that meant hay to feed sheep and goats, to feed Mongol babies, and it also meant hay to feed Mongol horses to ride into battle.
So Genghis Khan was someone at the right time and place, but he had no idea, and we didn't have any idea until this tree ring study of the last decade."
Another good essay from an OpenAI capabilities researcher I know (well, now former). I think focusing on control and abandoning alignment for now is is a reasonable strategy, but implementing it effectively will require an agreement not to RSI to ASI at max or close to max speed. Figuring out control is going to take time and money. Also, it only works up to a point, I think. In AI 2040 we talk about this.
Just been sent this of comments on a TikTok of me with monkeys. Regardless of my relationship with JJ, this is absolutely disgusting. It blows my mind how it’s 2023 and we’re seeing this shit. Racism ain’t funny, it’s time we all realise.
Steve Jobs said it best
Every single thing you see and hear was made by man. Your house, phone, music, car, roads, etc
NONE were any smarter than you.
You truly have the tools to change the world and do whatever tf you want 🤝