LessWrong 2.0 Reader

Can you give some examplse of something contained by my definition which you think shouldn't be considered exploitation?

The flinches aren't structureless particulars. Rather, they involve warping various perceptions. Those warped perceptions generalize a lot, causing other flaws to be hidden.

As a toy example, you could imagine someone attached to the idea of AI boxing. At first they say it's impossible to break out / trick you / know about the world / whatever. Then you convince them otherwise--that the AI can do RSI internally, and superhumanly solve computer hacking / protein folding / persuasion / etc. But they are attached to AI boxing. So they warp their perception, clamping "can an AI be very superhumanly capable" to "no". That clamping causes them to also not see the flaws in the plan "we'll deploy our AIs in a staged manner, see how they behave, and then recall them if they behave poorly", because they don't think RSI is feasible, they don't think extreme persuasion is feasible, etc.

A more real example is, say, people thinking of "structures for decision making", e.g. constitutions. You explain that these things, they are not reflectively stable. And now this person can't understand reflective stability in general, so they don't understand why steering vectors won't work, or why lesioning won't work, etc.

Another real but perhaps more controversial example: {detecting deception, retargeting the search, CoT monitoring, lesioning bad thoughts, basically anything using RL} all fail because creativity starts with illegible concomitants to legible reasoning.

(This post seems to be somewhat illegible, but if anyone wants to see more real examples of aspects of mind that people fail to remember, see https://tsvibt.blogspot.com/2023/03/the-fraught-voyage-of-aligned-novelty.html)

It's a question of resolution. Just looking at things for vibes is a pretty good way of filtering wheat from chaff, but you don't give scarce resources like jobs or grants to every grain of wheat that comes along. When I sit on a hiring committee, the discussions around the table are usually some mix of status markers and people having done the hard work of reading papers more or less carefully (this consuming time in greater-than-linear proportion to distance from your own fields of expertise). Usually (unless nepotism is involved) someone who has done that homework can wield more power than they otherwise would at that table, because people respect strong arguments and understand that status markers aren't everything.

Still, at the end of day, an Annals paper is an Annals paper. It's also true that to pass some of the early filters you either need (a) someone who speaks up strongly for you or (b) pass the status marker tests.

I am sometimes in a position these days of trying to bridge the academic status system and the Berkeley-centric AI safety status system, e.g. by arguing to a high status mathematician that someone with illegible (to them) status is actually approximately equivalent in "worthiness of being paid attention to" as someone they know with legible status. Small increases in legibility can have outsize effects in how easy my life is in those conversations. 
 

Otherwise it's entirely down to me putting social capital on the table ("you think I'm serious, I think this person is very serious"). I'm happy to do this and continue doing this, but it's not easily scalable, because it depends on my personal relationships.

Which one's do you see as the top ones?

To be clear, I am not arguing that evolution is an example of what I'm talking about. The analogy to thermodynamics in what I wrote is straightforwardly correct, no need to introduce KT-complexity and muddy the waters; what I am calling work is literally work. 

Somewhat interesting point, but perhaps could have been made much briefer.

The verbose writing style makes me wonder if an LLM was used in the writing process?

Human: "Look, can't you just be normal about this?"

GAA-optimized agent: "Actually-"

Hm, I guess this wouldn't work if the agent still learns an internalized RL methodology? Or would it? Say we have a base model, not much need for GAA because it's just doing token pred. We go into some sort of (distilled?) RL-based cot instruct tuning, GAA means it picks up abnormal rewards from the signal more slowly, ie. it doesn't do the classic boat-spinning-in-circles thing (good test?), but if it internalizes RL at some point its mesaoptimizer wouldn't be so limited, and that's a general technique so GAA wouldn't prevent it? Still, seems like a good first line of defense.

Haha! I really hope I don't have to start running everything I write through a slave Mentat to avoid avoidable errors. What a deontic double bind that'd be <3

I really like your post. Good how-to manuals like yours are rare and precious.

My current "background I" (maybe not the one from 2017, but one I would tend to deploy here in 2024) includes something like: "Kolmogorov complexity is a cool ideal, but it is formally uncomputable in theory unless you have a halting oracle laying around in your cardboard box in your garage labeled Time Travel Stuff, and Solomonoff Induction is not tractably approximably sampled by extant techniques that aren't just highly skilled MCMC".