Menu
Sign In Search Podcasts Charts People & Topics Add Podcast API Blog Pricing
Podcast Image

LessWrong (Curated & Popular)

"Responsible Scaling Policy v3" by HoldenKarnofsky

25 Feb 2026

Transcription

Chapter 1: What is the main topic discussed in this episode?

0.031 - 13.401 Holden Karnofsky

Responsible Scaling Policy v3 by Holden Karnofsky Published on February 24, 2026 All views are my own, not Anthropic's.

0

14.482 - 28.824 Unknown

This post assumes Anthropic's announcement of RSP version 3.0 as background. Today, Anthropic released its responsible scaling policy 3.0. The official announcement discusses the high-level thinking behind it.

0

Chapter 2: What are the original goals of Responsible Scaling Policies?

29.886 - 44.719 Unknown

This is a more detailed post giving my own takes on the update. First, the big picture. There's a list of bullet points here. I expect some people will be upset about the move away from a hard commitment forward slash binding ourselves to the master vibe.

0

45.425 - 57.019 Unknown

Anthropic has always had the ability to revise the RSP, and we've always had language in there specifically flagging that we might revise away key commitments in a situation where other AI developers aren't adhering to similar commitments.

0

58.06 - 68.833 Unknown

But it's been easy to get the impression that the RSP is binding ourselves to the mast and committing to unilaterally pause AI development and deployment under some conditions, and Anthropic is responsible for that.

0

Chapter 3: How has the implementation of RSPs gone so far?

70.18 - 83.582 Unknown

I take significant responsibility for this change. I have been pushing for this change for about a year now and have led the way in developing the new RSP. I am in favour of nearly everything about the changes we're making.

0

Chapter 4: What are the forcing functions for improved risk mitigations?

84.187 - 103.579 Unknown

I am excited about the roadmap, the risk reports, the move toward external review, and unwinding of some of the old requirements that I felt were distorting our safety efforts, more on all of this below. I think these changes are the right thing for reducing AI risk, both from Anthropic and from other companies if they make similar changes, as I hope they do.

0

104.96 - 124.015 Unknown

In my mind, this revision isn't being prompted by catastrophic risk from today's AI systems is now high, I don't think it is, or by, we've just realized that sufficient regulation isn't looking likely, I think this is not a very recent update. First and foremost, in my mind, it is about learning from design flaws and making improvements.

0

Chapter 5: What are the mixed success stories in AI safety?

125.378 - 140.659 Unknown

I always thought of the original RSP as a V1 that would be iterated on, and have been frustrated to see the extent to which it's been interpreted as a sacred cow or binding oneself to the mast such that revisions that go in a less self-binding direction are seen by many as inherently dissonable.

0

141.721 - 157.652 Unknown

I would have pushed for a very different, and far less ambitious, initial design if I'd thought this way about future changes. I generally think it's bad to create an environment that encourages people to be afraid of making mistakes, afraid of admitting mistakes and reticent to change things that aren't working.

0

158.673 - 179.712 Unknown

I think that dynamic currently applies somewhat to RSP-like policies, and I hope that changes. I'm not saying that I wish people shrugged off every revision with, hey, it's your policy, do what you want. I wish people simply evaluated whether the changes seem good on the merits, without starting from a strong presumption that the mere fact of changes is either a bad thing or a fine thing.

0

180.754 - 197.249 Unknown

It should be hard to change good policies for bad reasons, not hard to change all policies for any reason. I think a lot of people have a mentality like I worry that AI companies will do the wrong thing when it comes down to it, and I'm looking for them to bind their future actions as rigidly and tightly as possible.

0

198.03 - 207.124 Unknown

Policies that are vaguer and more flexible just leave more room for motivated reasoning. I think that is fair as far as it goes, but there's a flip side.

207.104 - 228.454 Unknown

The world, especially the world of AI, changes fast, and binding commitments about the future can leave you bound to things that aren't actually good for safety, and can make it hard to adapt to the situation as it actually stands and allocate resources effectively, a dynamic that risks poor prioritization, goodhotting, and general annoyance and backlash that I think one should care about quite a lot.

228.788 - 244.485 Unknown

I've always been aware of the latter issue and tried to be careful about what kinds of commitments are worth making, but I've updated toward thinking that binding commitments are harder to get right than I'd thought, while the kinds of policies that look frustratingly vague to many people can actually have a huge impact in the right context.

Chapter 6: How does RSP v3 aim to amplify positive outcomes and reduce negatives?

246.067 - 268.499 Unknown

I think a common attitude toward RSP version 3.0 is I reluctantly see why this is necessary, and I ultimately agree that the changes are needed and reasonable, but I'm sad about it. That's not my attitude. I am sad that we're in the political environment we're in, but taking this not terribly recent situation as a given, I am affirmatively excited about the new RSP.

0

269.54 - 289.828 Unknown

I think the old RSP did some things really well, but also had some perverse effects that don't seem widely understood and risked growing a lot as AI systems become more capable, more below. I think the new one is better and will ultimately be a more effective force for risk reduction. I don't think either could keep risks bound to low levels on a voluntary basis.

0

290.332 - 310.399 Unknown

I'm sure there are problems with this RSP too, and at some point there will likely be an RSP 4.0 with more changes, but I do think we're learning about what works and doesn't work, and these policies will hopefully have growing positives and shrinking negatives if we can treat them as works in progress rather than sacred cows. That's the end of the list.

0

310.419 - 318.55 Unknown

I think my viewpoint is probably easiest to understand via my story about the evolution and impacts, good and bad, of RSPs since the beginning.

0

Chapter 7: What concerns arise from moving away from unilateral commitments?

319.019 - 328.19 Unknown

This will cover what the original goals were, why we approached them the way we did initially, what went well with that approach, and what didn't, which motivates the changes we're making now.

0

329.332 - 330.974 Holden Karnofsky

After that is an FAQ section.

0

332.276 - 349.674 Unknown

Heading How it started The original goals of RSBs In 2023, I collaborated with META to develop and pitch the basic idea of responsible scaling policies. What we were trying to do is pretty well captured in the first blog PostMeter wrote about them.

0

Chapter 8: What messages does the RSP v3 send to regulators?

350.697 - 366.598 Unknown

Today I'd summarise the goals roughly as follows. Please read this as reporting my own thinking on the goals of RSBs, rather than Meters or anyone else's. Goal 1. Create forcing functions for AI developers to move with urgency and focus on risk mitigations.

0

367.719 - 381.78 Unknown

The idea was, if a company has a policy saying it isn't safe to train an AI model with X level of capabilities unless Y risk mitigations are in place, then hopefully the company is going to try very hard to get those risk mitigations in place.

0

381.76 - 401.608 Unknown

This doesn't rely on commitments being ironclad, only on something like it would be embarrassing to fall short of a standard the company has said it is trying to hit and associates with safe AI development and companies will work to avoid that kind of embarrassment. Goal 2. Create a testbed for practices and policies that can feed into policy frameworks.

0

402.145 - 419.969 Unknown

The idea was, if many major AI developers have adopted risk mitigation Y, and or agreed that risk mitigation Y is important for safety, then this will make it easier, both politically and practically, for regulation to require or nudge risk mitigation Y, or gate AI development or deployment on it.

0

421.071 - 432.69 Unknown

Again, this doesn't rely on commitments being ironclad, any given risk mitigation that is widely practiced and or publicly supported by industry at the moment can have a policy impact. A bit more specifics.

433.571 - 454.324 Unknown

I hope that if political will for AI risk reduction ended up strong, then voluntary practices and policies by companies would be taken as a floor for regulation, whereas if political will for AI risk reduction ended up weak, these might be the best we could get. Goal 3. Work toward consensus and common knowledge about AI risks and potential mitigations.

455.426 - 472.552 Unknown

There was already a lot of interest in evaluating AI systems for dual use to dangerous capabilities, but we hoped that responsible scaling policies would increase efforts to tie capabilities to threat models and generally improve the level of common knowledge about whether AI systems were becoming dangerous. not a core goal.

473.455 - 487.232 Unknown

Bring about a substantial, voluntary, non-regulation-backed pause in AI development. At the time, many people seemed to assume that this was a goal of ours and criticized the RSP effort on grounds that it seemed unrealistic.

487.651 - 509.143 Unknown

While I don't think it was or is totally implausible that a small number of AI developers could, if they were far ahead of all others, slow down AI development by some amount due to policies like this, it was never a major part of the hope, nor was it necessary to achieve the other goals listed above. Escape clauses. META's intro post on responsible scaling policies included this.

Comments

There are no comments yet.

Please log in to write the first comment.