what are your go-to intro resources to ai safety (ideally focusing on how to actually make agi not kill everyone, not just the prosaic stuff) for technical people? i feel my go-tos are getting a bit outdated
If I want to get something-like-my-worldview across, rather than technical info, I’ll usually send AI 2027 or AI 2040, and then follow up from there to see (1) if they jumped off the AI/AGI/ASI train at some point and (2) what more they want to learn
what are your go-to intro resources to ai safety (ideally focusing on how to actually make agi not kill everyone, not just the prosaic stuff) for technical people? i feel my go-tos are getting a bit outdated
The “Why Not Just...” sequence still seems good for what not to do/reconsidering certain things
What were your previous go-tos?
richard’s alignment from a deep learning perspective
every rob miles video ever
ai-2027 (and ai-2040)
Everyone who watched every Rob video ever should also read the lost Arbital Seuqnece, though it’s older.
https://intelligence.org/briefing/ still seems great to me
If I want to get something-like-my-worldview across, rather than technical info, I’ll usually send AI 2027 or AI 2040, and then follow up from there to see (1) if they jumped off the AI/AGI/ASI train at some point and (2) what more they want to learn