“Overall, my current view is that despite these uncertainties and trade-offs, we should try hard to put ourselves in a position to steer or restrain AI development when doing so is necessary for safety; that even limited or temporary success in this respect can make a significant positive difference; and that more thoroughgoing success remains, at least, a live possibility.”
Joe Carlsmith · March 19, 2026
Title and organization are as of the date said.
In context
“Overall, my current view is that despite these uncertainties and trade-offs, we should try hard to put ourselves in a position to steer or restrain AI development when doing so is necessary for safety; that even limited or temporary success in this respect can make a significant positive difference; and that more thoroughgoing success remains, at least, a live possibility.” He also says he disagrees with advocates who argue the safety community should give up on technical safety progress and pivot almost exclusively to capability restraint.
The speaker’s own caveats
However, I also think that safety-concerned advocates for capability restraint should acknowledge the genuine uncertainties and trade-offs it involves, and remain attentive to the possibility that as we learn more (both about alignment risk and about the broader situation), these trade-offs can tip the balance against available options for real-world capability restraint
Source
- On restraining AI development for the sake of safety · joecarlsmith.com (article)
- Archived copy (Wayback Machine)
- Content captured 2026-11-16, SHA-256 e2458e7bf6d41db5d1899a2a2c93bb9ed350961b872e2f35f677910cebb6dcb4
Cite this quote
Carlsmith, J. (2026, March 19). On restraining AI development for the sake of safety. joecarlsmith.com. https://joecarlsmith.com/2026/03/19/on-restraining-ai-development-for-the-sake-of-safety/
More from Joe Carlsmith
“Safety progress takes time. If you don’t have enough time, at each stage of AI development, to ensure the AIs that you build in the next stage won’t destroy humanity, then you’ll fail, and humanity will be destroyed. Automated alignment research helps a lot here, and that’s a core reason I think it’s so important (human alignment researchers are too scarce and slow). But even if AIs are doing most or all of the alignment work, it still matters how much time they have.”
Joe Carlsmith · March 19, 2026
Others on the same topic
“We will need to build the infrastructure to take advantage of this technology, but we will also need to go slowly and deliberately on developing the technology itself, to pace or otherwise control its development.”
Jack Clark, Co-founder, Anthropic · October 17, 2026
“For my part, I think the math and CS theory communities should absolutely hold the AI companies’ feet to the fire to behave better, both within our little domains, and (much more consequentially) in pacing AI progress, figuring out alignment, and preventing a catastrophe for our whole civilization.”
Scott Aaronson, Professor of Computer Science, University of Texas at Austin · October 9, 2026
“Yes, sir. I believe we should make it illegal and I do think it's possible.”
Daniel Kokotajlo, Lead, AI Futures Project · September 30, 2026
All quotes on pause →All quotes on race dynamics →All quotes on regulation →