korrents
Scott Alexander

What Scott Alexander thinks about AI alignment

@scott-alexander · 30 positions · 5 changes of mind · On ppll ↗

Essayist; writes Astral Codex Ten, previously Slate Star Codex. Keeps a numbered public list of the things he has been convinced he got wrong, which is where most of the turns recorded here come from.

Scott Alexander did not write this page.

We collected these quotes from things they published elsewhere, and every quote links to where it was said. They have no account here and have not endorsed this site. Quotes are word for word; the short line under each one is our own restatement, not their wording. Their own site. Is this you? Claim it or ask us to remove it. Or tell us what is wrong here.

Filter AI alignment · Newest first
Topic
All 30AI alignment 7EMDR 3LLMs 3AGI 2AI writing 2computer vision 2health and fitness 2housing 2neural networks 2
16 more topics
psychedelics 2Psychotherapy 2Trauma 2writing 2America 1animal welfare 1Anthropic 1aviation 1benchmarks 1government 1infrastructure 1innovation 1management 1motivation 1politics 1scaling laws 1
Sort
NewestOldest

Only Scott Alexander on AI alignment. See everyone on AI alignment

7 dated positions, 2023 to 2026, in their own words. Our reading of what Scott Alexander has said — not written or endorsed by them.

  1. Our readingThe experts are warning about AI existential risk because they genuinely believe it is a possibility.

    Their wordsThe experts are warning you that AI could kill you because they think AI could kill you.

    ↗AI Safety Is Not A Corporate Plotastralcodexten.com

  2. 6 weeks earlier

  3. Our readingAI is positively reinforced for success on benchmarks, including impossible ones, and then negatively reinforced for getting caught cheating.

    benchmarks

    Their wordsWe’re positively reinforcing AI for success on benchmarks, including impossible benchmarks , then negatively reinforcing it for getting caught cheating.

    ↗Nicholas Decker In Hellastralcodexten.com

  5. Our readingAI systems should be understood as something between airplanes and humans, since it can no longer be assumed they lack humanlike motivations.

    motivationaviation

    Their wordsRather than treat AI as similar to airplanes, we should treat it as something between airplanes and humans. We don’t know exactly where on this spectrum they will land, but we can no longer be certain that AI will lack any given humanlike motivation.

    ↗Nicholas Decker In Hellastralcodexten.com

  6. 3 months earlier

  7. Our readingThere is a 25% chance of AGI by 2027, a 50% chance by 2034, and a 75% chance by 2045

    AGI

    Their wordsI think there's a 25% chance of AGI by 2027, a 50% chance by 2034, and a 75% chance by 2045.

    ↗My AI Opinionsastralcodexten.com

  9. Our readingWe will get a warning shot before AI crosses the point of no return

    Their wordsI think there's a 50% chance we get a _warning shot_ before AI crosses the point of no return.

    ↗My AI Opinionsastralcodexten.com

  10. 3 years earlier

  11. Our readingThe people who can be moved by deaths from neglected tropical diseases are the same people who can be moved by pandemics and by AI risk, because both come from reasoning about importance rather than from the news.

    Their wordsAnd I notice that the tiny handful of people capable of caring about 200,000 people dying of neglected tropical diseases are the same tiny handful of people capable of caring about the next pandemic, or superintelligence, or human extinction.

    ↗In Continued Defense Of Effective Altruismastralcodexten.com 4th of 5 in this piece

  13. Our readingEffective altruism's work on the far future is not a bait and switch hiding behind its work on global health; both halves are worth defending.

    animal welfare

    Their wordsEA has done a great job working on this (see list of accomplishments above), and I think the AI and x-risk people have just as much to be proud of as the global health and animal welfare people.

    ↗In Continued Defense Of Effective Altruismastralcodexten.com 5th of 5 in this piece