Rossin

Replying toAnnouncing the Inverse Scaling Prize ($250k Prize Pool)

Announcing the Inverse Scaling Prize ($250k Prize Pool)

Would the ability to deceive humans when specifically prompted to do so be considered an example? I would think that large LMs get better at devising false stories about the real world that people could not distinguish from true stories.

Replying toOn A List of Lethalities

Rossin4y

On A List of Lethalities

On the idea of "we can't just choose not to build AGI". It seems like much of the concern here is predicated on the idea that so many actors are not taking safety seriously, so someone will inevitably build AGI when the technology has advanced sufficiently.

I wonder if struggles with AIs that are strong enough to cause a disaster but not strong enough to win instantly may change this perception? I can imagine there being very little gap if any between those two types of AI if there is a hard takeoff, but to me it seems quite possible for there be some time at that stage. Some sort of small/moderate disaster... (read more)

Replying toFixed points and free will

Rossin4y

Fixed points and free will

I feel like a lot of the angst about free will boils down to conflicting intuitions.

It seems like we live in a universe of cause and effect, thus all my actions/choices are caused by past events.
It feels like I get to actually make choices, so 1. obviously can't be right.

The way to reconcile these intuitions is to recognize that yes, all the decisions you make are in a sense predetermined, but a lot of what is determining those decisions is who you are and what sort of thing you would do in a particular circumstance. You are making decisions, that experience is not invalidated by a fully deterministic universe. It's just that you are who you are and you'll make the decision that you would make.

That’s true, there was a huge amount of outrage even before those details came out however.

Replying toThe Efficient LessWrong Hypothesis - Stock Investing Competition

Rossin4y

The Efficient LessWrong Hypothesis - Stock Investing Competition

I of course don’t have insider information. My stance is something close to Buffett’s advice “be fearful when others are greedy, and greedy when others are fearful”. I interpret that as basically that markets tend to be overly reactionary and if you go by fundamentals representing the value of a stock you can potentially outperform the market in the long run. To your questions, yes disaster may really occur, but my opinion is that these risks are not sufficient to pass up the value here. I’ll also note that Charlie munger has been acquiring a substantial stake in BABA, which makes me more confident in its value at its current price.

Replying toThe Efficient LessWrong Hypothesis - Stock Investing Competition

Rossin4y

The Efficient LessWrong Hypothesis - Stock Investing Competition

Alibaba (BABA) - the stock price has been pulled down by fear about regulation, delisting, and most recently instability in China as it's zero covid policy fails. However, as far as I can tell, the price is insanely low for the amount of revenue Alibaba generates and the market share that it holds in China.

Current bioethics norms will strongly condemn this sort of research, which may make it challenging to pursue in the nearish term. The consensus is strongly against, which will make acquiring funding difficult and any human CRISPR editing is completely off the table for now. For example, He Jiankui CRISPR edited some babies in China to make them less susceptible to HIV and went to prison for it.

Replying to Strategies for keeping AIs narrow in the short term

Rossin4y

Strategies for keeping AIs narrow in the short term

Do I understand you correctly as endorsing something like: it doesn’t matter how narrow an optimization process is, if it becomes powerful enough and is not well aligned, it still ends in disaster

Strategies for keeping AIs narrow in the short term

Rossin

Disclaimer: I don’t have any particular expertise in AI safety, I just had some thoughts and this seemed like the place to put them.

The bleak outlook that Eliezer presents here and elsewhere seems to in large part be driven by the expectation that AGI will be developed before the technical problem of how to robustly align it is solved, and that poorly aligned AGIs tend to destroy the world.

One way to potentially get more time to solve the technical problem is to expend more effort on activism with the goal of convincing the groups that are trying to create increasingly advanced AIs to slow down/stop and at least not attempt to create... (read 877 more words →)

Replying toWhat I Was Thinking About Before Alignment

Rossin4y

What I Was Thinking About Before Alignment

I’m not sure the problem in biology is decoding. At least not in the same sense it is with neural networks. I see the main difficulty in biology more one of mechanistic inference where a major roadblock may be getting better measurements of what is going on in cells over time rather some algorithm that’s just going to be able to overcome the fact that you’re getting both very high levels of molecular noise in biological data and single snapshots in time that are difficult to place in context. With a neural network you have the parameters and it seems reasonable to say you just need some math to make it more... (read more)

Replying toAre there substantial research efforts towards aligning narrow AIs?

Rossin4y

Are there substantial research efforts towards aligning narrow AIs?

In general the observation from working in the field is that if you have a simple metric, people will figure out how to game it. So you need to build in a lot of safeguards, and you need to evolve all the time as the spammers/abusers evolve. There's no end point, no place where you think you're done, just an ever changing competition.

That's what I was trying to point at in regards to the problem not being patchable. It doesn't seem like there is some simple patch you can write, and then be done. A solution that would work more permanently seems to have some of the "impossible" character of AGI alignment and trying to solve it on that level seems like it could be valuable for AGI alignment researchers.

Are there substantial research efforts towards aligning narrow AIs?

Rossin

I am certainly not an expert, but what I have read in the field of AI alignment is almost solely targeted at handling the question of how to make sure that a general AI does not start optimizing for outcomes that humans do not want to happen.

This is, to me, an obviously important problem, in that it seems like we only get one chance at it and if we fail we are all Doomed.

However, it also seems like a very difficult problem to get good feedback on since there are no AGIs to test different strategies on, which can make estimating the promise of different avenues of research very difficult.

A related problem... (read 332 more words →)

The Problem with Giving Advice

Rossin

Epistemic Status: I'm highly confident this is a phenomenon that occurs with a lot of advice people give, but I'm quite uncertain about the best way to deal with it when trying to give advice to more than one person.

The main thing people fail to consider when giving advice is that someone with ostensibly the same problem may require a vastly different solution than themselves. The underlying cause of the problem in many cases may be the exact opposite of what it was for you.

Many issues in life are caused by being too extreme. Either extreme is problematic-both too much of something or too little. Often people only think about one extreme... (read 508 more words →)

Laughing Away the Little Miseries

Rossin

Note: This falls under the category of self-help advice, the usefulness of which tends to be variable between individuals. This technique has noticeably improved my day to day feelings of well being, but people are often very different and your mileage may vary.

It is notoriously difficult to increase your long term happiness. The psychology literature has coined the term ‘happiness set point’ to describe the fact that people seem to stay around the same level of happiness despite major events in their lives that one might naively expect to make them more or less happy. The classic examples of this effect are that lottery winners aren’t nearly as happy as they thought... (read 493 more words →)

The Strengths of the Two Systems of Cognition

Rossin

The two systems of cognition a la Kahneman each seem to each have roles that they fulfill better than the other and in which, for optimal performance, the other system ought not to interfere. Until recently, I had only really thought about when System 2 ought to override System 1, but I believe other cases are worth considering as well.

It is important to be able to get System 2 to shut up and let System 1 work in certain circumstances. Physical movement is an excellent example of this. When you shoot a basketball, throw a football, or use a video game controller you may notice that when you are “on fire” you

... (read 680 more words →)

The Balance Between Hard Work and Exhaustion

Rossin

Rationalists often find difficult, important challenges to work on and they become very excited and passionate about their causes. I expect it is common (because it happened to me and I have heard references to similar episodes by others) that such causes seem so important that aspiring rationalists set unreasonably high standards of dedication for themselves.

I think the concept of giving an extraordinary effort is a very important one, your brain wants to be lazy and if you are trying to do something challenging that you think is vitally important you want to push yourself to actually work hard, rather than merely “work hard”. A significant portion of this effort is exploring

... (read 663 more words →)

Positive Focusing

Rossin

Gendlin’s technique of Focusing primarily focuses (hehe) on problems or negative felt senses. Something I have not seen discussed much is that one can apply the concepts of Focusing to many different felt senses that are not problems, or even negative in any way.

At least in my experience, you can equally investigate good feelings with Focusing as you can problems. And, much as Focusing on a problem improves your understanding of that problem, Focusing on a positive felt sense can improve your understanding of what makes you feel good. By repeatedly investigating feeling good in this way, I think it makes it much easier to piece together an accurate picture of what

... (read 624 more words →)

Strategic Goal Pursuit and Daily Schedules

Rossin

In the post Humans Are Not Automatically Strategic, Anna Salamon writes:

there are clearly also heuristics that would be useful to goal-achievement (or that would be part of what it means to “have goals” at all) that we do not automatically carry out. We do not automatically:
(a) Ask ourselves what we’re trying to achieve;
(b) Ask ourselves how we could tell if we achieved it (“what does it look like to be a good comedian?”) and how we can track progress;
(c) Find ourselves strongly, intrinsically curious about information that would help us achieve our goal;
(d) Gather that information (e.g., by asking as how folks commonly achieve our goal, or similar

... (read 1533 more words →)

LESSWRONG
LW

LESSWRONG
LW

Positive Focusing

The Balance Between Hard Work and Exhaustion

The Problem with Giving Advice

Laughing Away the Little Miseries

Rossin

Rossin

Strategies for keeping AIs narrow in the short term

Are there substantial research efforts towards aligning narrow AIs?

The Problem with Giving Advice

Laughing Away the Little Miseries

The Strengths of the Two Systems of Cognition

The Balance Between Hard Work and Exhaustion

Positive Focusing

Rossin

Positive Focusing

The Balance Between Hard Work and Exhaustion

The Problem with Giving Advice

Laughing Away the Little Miseries

Rossin

Rossin

Strategies for keeping AIs narrow in the short term

Are there substantial research efforts towards aligning narrow AIs?

The Problem with Giving Advice

Laughing Away the Little Miseries

The Strengths of the Two Systems of Cognition

The Balance Between Hard Work and Exhaustion

Positive Focusing