Boaz Barak
Article
Boaz Barak is a recurring person in the Astral Codex Ten archive, appearing 3 times across 3 issues between January 30, 2024 and March 01, 2026. The archive places it in contexts such as “an experiment courtesy of Metaculus, Scott Aaronson, and Boaz Barak”; “Boaz Barak (friend of Scott Aaronson’s, now working on OpenAI alignment team) has six thoughts on AI safety”; “Boaz Barak suggests this is the case”. It most often appears alongside Anthropic, James Grugett, Metaculus.
Metadata
- Category: People
- Mention count: 3
- Issue count: 3
- First seen: January 30, 2024
- Last seen: March 01, 2026
Appears In
Related Pages
-
- Anthropic (2 shared issues)
-
- James Grugett (2 shared issues)
-
- Metaculus (2 shared issues)
-
- Musk (2 shared issues)
-
- Nate Silver (2 shared issues)
-
- OpenAI (2 shared issues)
-
- PEPFAR (2 shared issues)
-
- Sam Altman (2 shared issues)
-
- Scott Aaronson (2 shared issues)
-
- Trump (2 shared issues)
-
- Twitter (2 shared issues)
-
- Ukraine (2 shared issues)
External Links
Source Context
Recovered passages from the original issue text. When the raw archive preserved outbound links inside the source passage, they are listed directly under the quote.
Here’s an experiment courtesy of Metaculus, Scott Aaronson, and Boaz Barak. Aaronson and Barak wrote a blog post trying to divide AI scenarios into five categories, which Metaculus summarizes as:
36: Boaz Barak (friend of Scott Aaronson’s, now working on OpenAI alignment team) has six thoughts on AI safety. It’s all pretty moderate and thoughtful stuff - what I find interesting about it is that the acknowledgments say Sam Altman provided feedback (although “do[es] not necessarily endorse any of its views”). I think this is a useful window into OpenAI’s current alignment thinking, or at least into the fact that they currently have alignment thinking. Not much to complain about in terms of specifics and glad people like Boaz are involved.
Inline links: six thoughts on AI safety
Overall: We can’t see how any of OpenAI’s claimed methods for enforcing their red lines would work except possibly if they’re allowed to implement technical safeguards that block certain lawful use, which they’ve shared so little about that we can’t evaluate it. Boaz Barak suggests this is the case. If this is right, it’s strange that they don’t elsewhere stress this as the linchpin of their approach, or show the part of the agreement that guarantees them this ability. Further clarification on this point would be very helpful.
Inline links: suggests