See also this summary + commentary on METR's review: https://www.lesswrong.com/posts/wPQaeMvHXq9Ee2tEm/summary-and-comments-on-anthropic-s-pilot-sabotage-risk

That's right. We did some followup experiments doing the head-to-head comparison: the tools seem to speed up the contractors by 2x for the weak adversarial examples they were finding (and anecdotally speed us up a lot more when we use them to find more egregious failures). See https://www.lesswrong.com/posts/n3LAgnHg6ashQK3fF/takeaways-from-our-robust-injury-classifier-project-redwood#Quick_followup_results; an updated arXiv paper with those experiments is appearing on Monday. 

The main reason is that we think we can learn faster in simpler toy settings for now, so we're doing that first. Implementing all the changes I described (particularly changing the task definition and switching to fine-tuning the generator) would basically mean starting over from scratch anyway.

Indeed. (Well, holding the quality degradation fixed, which causes a small change in the threshold.)

I read Oli's comment as referring to the 2.4% -> 0.002% failure rate improvement from filtering.

Yeah, I think that might have been wise for this project, although the ROC plot suggests that the classifiers don't differ much in performance even at noticeably higher thresholds.

For future projects, I think I'm most excited about confronting the problem directly by building techniques that can succeed in sampling errors even when they're extremely rare.

Excellent question -- I wish we had included more of an answer to this in the post.

I think we made some real progress on the defense side -- but I 100% was hoping for more and agree we have a long way to go.

I think the classifier is quite robust in an absolute sense, at least compared to normal ML models. We haven't actually tried it on the final classifier, but my guess is it takes at least several hours to find a crisp failure unassisted (whereas almost all ML models you can find are trivially breakable). We're interested in people giving it a shot! :)

Part of what's going on here is that IMO we made more progress on the offense side than the defense side. We didn't measure that as rigorously, but it seemed like the token substitution tool speeds up the attack by something like 10x. I think the attack tools are one of the parts of the project I'm most excited about, and we might push even harder on them in upcoming projects.

I think you're correct that the amount of improvement was pretty limited by capabilities. 4,900 labels (for tool-assisted adversarial examples) just aren't many for a 304M-parameter model. That's one reason we don't find it totally disturbing that the offense is winning for now; with much bigger models sample efficiency should increase dramatically, and the baseline cost of constructing the model will increase as well, making it comparatively cheaper to spend more on adversarial training.

That said, this was just our first go and I think we can make a lot more progress right away. It'll help to have a crisper task definition that doesn't demand that the model understand thousands of possible ways injury could occur. It'll help to have attacks that can produce higher volumes of data and cover a larger portion of the space. I don't think we know how to really kill it yet, but I'm excited to keep working on it.