LessWrong 2.0 Reader

Obviously keep working, but stop talking where people who are trying to destroy the world can hear. If you're Neel Nanda and work for a company trying to destroy the world, consider not publishing anything else at all, and only publishing useless versions of your work, because your work being useful for resilient moral alignment depends on a long chain of things that publishing it makes near impossible.

EY may be too busy to respond, but you can probably feel pretty safe consulting with MIRI employees in general. Perhaps also Conjecture employees, and Redwood Research employees, if you read and agree with their views on safety. That at least gives you a wider net of people to potentially give you feedback.

The problem is that "helpful" oracle AI will not stay helpful for long, if there is any incentive to produce things which are less helpful. Your beliefs are apparently out of date: we have helpful AI now, so that's an existence disproof of "helpful ai is impossible". But the threat of AI being more evolutionarily fit, and possibly an AI taking sudden and intense action to make use of its being more evolutionarily fit, is still hanging over our heads; and it only takes one hyperdesperate not-what-you-meant seeker.

Concretely, I think your posts are in fact a great (but not at all worst-case) example of things that have more cost than benefit, and I think you should keep working but only talk to people in DMs. Time is very, very short, and if you accidentally have a pivotally negative impact, you could be the one that burns the last two days before the world is destroyed.

I would give examples of things that shouldn't have been published and are why I agree, but that would be missing the point, wouldn't it?

Let's put it this way: I think most "alignment" or "safety" research is in fact nothing of the kind, and most people responding are deluding themselves so as to avoid having to consider the possibility of needing to go back to the drawing board.

As usual, capability (ability to figure out things about ai) generalizes further than alignment (ability to aim your ability to understand ai at actually making your knowledge produce utilitarian(-prioritarian)-morally-good outcomes).

Did you ever see any plane that far? I saw only planes above me (10 km) and they are almost like dots. 

The difference between optics and radar is that with optics you need to know where to look - but the radar has constant 360 perception. 

I'm skeptical of the RLHF example (see also this post by Paul on the topic [LW · GW]).

That said, I agree that if indeed safety researchers produce (highly counterfactual) research advances that are much more effective at increasing the profitability and capability of AIs than the research advances done by people directly optimizing for profitability and capability, then safety researchers could substantially speed up timelines. (In other words, if safety targeted research is better at profit and capabilities than research which is directly targeted at these aims.)

I dispute this being true.

(I do think it's plausible that safety interested people have historically substantially advanced timelines (and might continue to do so to some extent now), but not via doing research targeted at improving safety, by just directly doing capabilities research for various reasons.)

I think pain is a little bit different than that. It's the contrast between the current state and the goal state. This constrast motivates the agent to act, when the pain of contrast becomes bigger than the (predicted) pain of acting.

As a human, you can decrase your pain by thinking that everything will be okay, or you can increase your pain by doubting the process. But it is unlikely that you will allow yourself to stop hurting, because your brain fears that a lack of suffering would result in a lack of progress (some wise people contest this, claiming that wu wei is correct).

Another way you can increase your pain is by focusing more on the goal you want to achieve, sort of irritating/torturing yourself with the fact that the goal isn't achieved, to which your brain will respond by increasing the pain felt by the contrast, urging action.

Do you see how this differs slightly from your definition? Chronic pain is not a continuous reduction in agency, but a continuous contrast between a bad state and a good state, which makes one feel pain which motivates them to solve it (exercise, surgery, resting, looking for painkillers, etc). This generalizes to other negative feelings, for instance to hunger, which exists with the purpose to be less pleasant than the search for food is, such that you seek food.

I warn you that avoiding negative emotions can lead to stagnation, since suffering leads to growth (unless we start wireheading, and making the avoidance of pain our new goal, because then we might seek hedonic pleasures and intoxicants)

Some features I'd like:

a 'mark read' button next to posts so I could easily mark as read posts that I've read elsewhere (e.g. ones cross-posted from a blog I follow)

a 'not interested' button which would stop a given post from appearing in my latest or recommended lists. Ideally, this would also update my recommended posts so as to recommend fewer posts like that to me.

a 'read later' button which will put the post into a reading list for me that I can come back to later.

a toggle button for 'show all' / 'show only unread' so that I could easily switch between the two modes.

These features would help me keep my 'front page' feeling cleaner and more focused.

You are of course perfectly right. What I meant was: so that their convex hull is full-dimensional and contains the origin. I fixed it. Thanks for spotting this!

oh I see, by all(sensor_preds) I meant  sum([logit_i] for i in n_sensors) (the probability that all sensors are activated). Makes sense, thanks!