ben_w

6 ms·

31.
by ben_w 5d ago
> Also no idea what viruses and cancers have to do with this. Cancer is surely very poor reproductively because they never spread to other hosts The host's survival is not the benchmark of the reproductive unit, which are the cancer
32.
by ben_w 6d ago
Benchmaxxing is the default case, and always has been. It's really, really difficult to avoid it even when you care to stop yourself; and it's not even just a problem in machine learning, it's the standard failure mode of all
33.
by ben_w 6d ago
Normally, businesses spend a lot of money saying "no, our products are 100% safe" and get very upset with people who say they are not. You are literally looking at someone who resigned because the company is acting unsafely, where
34.
by ben_w 6d ago
> First I've heard of this, in what case was free speech used to justify a right to show a specific ad? All of them? This is pretty much the default case. If Meta or Google or whoever say "we don't want to show this ad&quo
35.
by ben_w 7d ago
Ryanair are the only ones that I have flown with who often depart roughly when they were supposed to land.
36.
by ben_w 7d ago
> Last I checked usernames aren't a public good. You're part of a social network and they get to set the policy. I'd count this as "strong-armed into this". The giants want all the benefits of being public spaces
37.
by ben_w 7d ago
> Yes, Ryanair stole 100 euro from me once, but if next flight is 50 to 250 euro cheaper than competition... I swore off them a few years ago. Plane started boarding around when it was due to land. We got on, seeing about 3cm of snow aro
38.
by ben_w 7d ago
> Is there something more we need to understand? Only if you're interested in the specific failure modes that LLMs have. That's all this story is.
39.
by ben_w 7d ago
If we were actively trying to make this "win" in the Stockfish sense, it would likely be 99%. We are trying to make a system that doesn't want to "win" in the sense, but wants to "win" by being helpful, ha
40.
by ben_w 7d ago
> I’ve yet to see a rational argument for how we go from super intelligent LLMs to human extinction or extermination. Much the same way our ancestors went from a super intelligent primate to this: https://en.wikipedia.org&#x2F
41.
by ben_w 7d ago
Same way you get "more than 10%" prediction on "I don't know what moves Stockfish will make when I play against it, but I know I will lose". In fact, I will lose in part because I don't know what moves Stockf
42.
by ben_w 7d ago
I mean much more generally than doom. It's like, Altman's reputation in general is not great, and that includes people who've worked there.
43.
by ben_w 7d ago
Sure, sure. There's many others like this to help you combine whatever you do feel you can put a number to. For me, that particular question is "probably 0, but with 100% variance". This is because I think most of the things
44.
by ben_w 7d ago
> The current government is maximally evil and interested in using technology to keep power and crush dissent. You don't need to introduce hypotheticals! No no, I assure you, it's quite capable of getting very much worse.
45.
by ben_w 7d ago
A flat distribution is still a distribution, and correct exploration would have revealed that the distribution is flat. The agent appears to have gained the false belief that it has learned something and done some exploring, when in fact it
46.
by ben_w 7d ago
Mine is around 10%. There's lots of moving parts and we all have to input our best-guesses as to how they interact. Some are predictable (e.g. "military will want capabilities, want them able to choose targets"). Others are n
47.
by ben_w 7d ago
The paperclip maximizer is also at the intentional direction of a human or humans. It's not "AI surprises everyone by having a thing for paperclips", it is "idiot tells AI to maximise paperclips no matter what, and then
48.
by ben_w 7d ago
I'm old enough to remember when people dismissed all the doom coming from these companies as "marketing". (A thing many of them have been entirely consistent about since GPT-2, or indeed earlier given the founding documents).
49.
by ben_w 7d ago
Talk is indeed cheap. Thing is, while there are many historical examples of a highly organised bunch of people with a lot of resources who engage in campaigns of violence for various reasons, many such attempts fail disastrously. Furthermor
50.
by ben_w 7d ago
There's a lot more than one chip fab. A while back Yudkowsky wrote that a ban would only work if was enforced by airstrikes. By a game of telephone, some people read "bomb", but there's a very big difference between &quo
51.
by ben_w 7d ago
I think you misunderstood my point, just as the other commentor misunderstood one level up. Perhaps a different approach to explain the problem here: "It ain't what they don't know, it's what they know for sure that just
52.
by ben_w 7d ago
> Apropos of TFA, if one is to believe LLMs can carve a threatening path to humanity end, how could one not see the same for our much older extant threats? Which "much older" extant threats? Global warming and nukes are industr
53.
by ben_w 7d ago
> You believe in AGI, but not ASI then? I have observed that these initialisms do not have generally agreed upon meanings; they are not useful for discussions with people who have not already stated what precisely they count. There is di
54.
by ben_w 7d ago
> I don’t think ai models come close to nuclear weapons or to run-of-the-mill, everyday carbon emissions in terms of danger to humanity. Even today, AI is the biggest discrete threat to bringing carbon emissions under control. Almost eve
55.
by ben_w 8d ago
I think that mainly changes the nature of the threat rather than anything else. Even if we didn't already have any wealth inequality or consolidated power, they're already starting to become capable of creating those things for th
56.
by ben_w 8d ago
LLMs can already control robots. Today's LLMs are inept at it, but they can do it. Things will be bad, unless someone does something to stop it. Will anyone do that? We don't know. https://xkcd.com/2278/
57.
by ben_w 8d ago
I suspect you two are discussing different AI. I think you (and I, and everyone who wants AI development to pause while we catch up with the implications at least) is looking where the ball is going. I have the general (non specific to anyo
58.
by ben_w 8d ago
Yes, but Venus has 93 times the atmospheric pressure of Earth, and it's overwhelmingly CO2. Earth isn't likely to get a true runaway like that until the sun gets another billion or so years on the clock, and when it does it will
59.
by ben_w 8d ago
While true, the impact from global warming is likely to be less crops rather than no crops.
60.
by ben_w 8d ago
All human behaviour since no later than the Polynesians turned the Pacific islands into a network of sailing trade routes. "Humans move" is one of the easiest things to predict. The biggest problem today is that they generally mov

More ›