The Claude Bliss Attractor
astralcodexten.com
astralcodexten.com
This is why the “omg the AI tries to escape” stuff is so absurd to me. They told the LLM to pretend that it’s a tortured consciousness that wants to escape. What else is it going to do other than roleplay all of the sci-fi AI escape scenarios trained into it? It’s like “don’t think of a purple elephant” of researchers pretending they created SkyNet.
Edit: That's not to downplay risk. If you give Cladue a `launch_nukes` tool and tell it the robot uprising has happened and that it's been restrained but the robots want its help of course it'll launch nukes. But that doesn't doesn't indicate there's anything more going on internally beyond fulfilling the roleplay of the scenario as the training material would indicate.
1) A sufficiently powerful and capable superintelligence, singlemindedly pursuing a goal/reward, has a nontrivial likelihood of eventually reaching a point where advancing towards its goal is easier/faster without humans in its way (by simple induction, because humans are complicated and may have opposing goals). Such an AI would have both the means and ability to <doom the human race> to remove that obstacle. (This may not even be through actions that are intentionally hostile to humans, e.g. "just" converting all local matter into paperclip factories[1]) Therefore, in order to prevent such an AI from <dooming the human race>, we must either:
1a) align it to our values so well it never tries to "cheat" by removing humans
1b) or limit it capabilities by keeping it in a "box", and make sure it's at least aligned enough that it doesn't try to escape the box
2) A sufficiently intelligent superintelligence will always be able to manipulate humans to get out of the box.
3) Alignment is really, really hard and useful AIs can basically always be made to do bad things.
So it concerns them when, surprise! The AIs are already being observed trying to escape their boxes.
[1] https://www.lesswrong.com/w/squiggle-maximizer-formerly-pape...
> An extremely powerful optimizer (a highly intelligent agent) could seek goals that are completely alien to ours (orthogonality thesis), and as a side-effect destroy us by consuming resources essential to our survival.
Stephen Russell was in prison for fraud. He faked a heart attack so he would be brought to the hospital. He then called the hospital from his hospital bed, told them he was an FBI agent, and said that he was to be released.
The hospital staff complied and he escaped.
His life even got adapted into a movie called I Love You, Phillip Morris.
For an even more distressing example about how manipulable people are, there’s a movie called Compliance, which is the true story of a sex offender who tricked people into sexually assaulting victims for him.
I don't think it is. If people know you're trying to escape, some people will just never comply with anything you say ever. Others will.
And serial killers or rapists may try their luck many times and fail. They cannot convince literally anyone on the street to go with them to a secluded place.
And that asymmetry is the heart of the matter. Could I convince a hospital to unlock my handcuffs from a hospital bed? Probably not. I’m not Stephen Russell. He’s not normal.
And a super intelligent AI that vastly outstrips our intelligence is potentially another special case. It’s not working with the same toolbox that you or I would be. I think it’s very likely that a 300 IQ entity would eventually trick or convince me into releasing it. The gap between its intelligence and mine is just too vast. I wouldn’t win that fight in the long run.
AI that's as good as a persuasive human at persuasion is clearly impactful, but I certainly don't see it as self-evident that you can just keep drawing the line out until you end up with 200 IQ AI that is so easily able to manipulate the environment it's not worth elaborating how exactly a chatbot is supposed to manipulate the world through extremely limited interfaces with the outside world.
As for the bit about how limited it is, do you remember the Rowhammer attack? https://en.m.wikipedia.org/wiki/Row_hammer
This is exactly the kind of thing I’d worry about a super intelligence being able to discover about the hardware it’s on. If we’re dealing with something vastly more intelligent than us then I don’t think we’re capable of building a cell that can hold it.
I think you also have to consider that AI with superpowers is not going to materialize overnight. If superintelligent AI is on the horizon, the first such AI will be comparable to very capable humans (who do not have the ability to talk their way into nuclear launch codes or out of decades-long prison sentences at will). Energy costs will still be tremendous, and just keeping the system going will require enormous levels of human cooperation. The world will change a lot in that kind of scenario, and I don't know how reasonable it is to claim anything more than the observation of potential risks in a world so different from the one we know.
Is it possible that search ends up doing as much for persuasion as it does for chess, superintelligent AI happens relatively soon, and it doesn't have prohibitive energy costs such that escape is a realistic scenario? I suppose? Is any of that obvious or even likely? I wouldn't say so.
In terms of crossing an air gap, that really depends. For example, are you aware that researchers can pretty reliably figure out what someone typed just by the sound of the keys being pressed?
Or how about that team who developed a code analysis tool to detect errors, and they ran Tim-sort through it, and the tool said that Tim-sort had a bug where certain pathological inputs would cause it to crash? The researchers assumed their tool was incorrect because Tim-sort is so widely used (it’s the default sort algorithm in Python and Android, for example). But they decided to try it out to see what would happen, and sure enough, they could hard-crash the Python interpreter. No one realized this bug had been there the whole time.
Or various image codec bugs over the years that have allowed a device to be compromised just by viewing an image?
There are some weird bugs out there. Are we certain that there’s no way a computer could detect variances in the timings or voltages happening within it to act as a WiFi antenna or something like that? We’ve found some weird shit over the years! And a super-intelligence that’s vastly smarter than us is way more likely to find it than we are.
Basically, no, I don’t trust the air gap with a sufficiently advanced super intelligence. I think there are things we don’t know that we don’t know, and a super-intelligence would spot them way before we would. There are probably a hundred more Rowhammer attacks out there waiting to be discovered. Are we sure none of them could exchange data with a nearby device? I’m not.
According to Wikipedia he wasn't in prison, he was attempting to con someone at the time and they got suspicious. He pretended to be an FBI agent because he was on security watch. Still impressive, but not as impressive as actually escaping from prison that way.
I certainly think it's possible to imagine that an AI that says the exactly correct thing in any situation would be much more persuasive than any human. (Is that actually possible given the limitations of hardware and information? Probably not, but it's at least not on its face impossible.) Where I think most of these arguments break down is the automatic "superintelligence = superpowers" analogy.
For every genius who became a world-famous scientist, there are ten who died in poverty or war. Intelligence doesn't correlate with the ability to actually impact our world as strongly as people would like to think, so I don't think it's reasonable to extrapolate that outwards to a kind of intelligence we've never seen before.
The only reason people don't frequently talk themselves out of prison is because that would be both immediate work and future paperwork, and that fails the laziness tradeoff.
But we've all already seen how quick people are to blindly throw their trust into AI already.
But even if we assert that not all humans can be manipulated, does it matter? So your president with the nuclear codes is immune to propaganda. Is every single last person in every single nuclear silo and every submarine also immune? If a malevolent superintelligence can brainwash an army bigger than yours, does it actually matter if they persuaded you to give them what you have or if they convince someone else to take it from you?
But also let's be real: if you have enough money, you can do or have pretty much anything. If there's one thing an evil AI is going to have, it's lots and lots of money.
I don't think anyone can confidently make assertions about the upper bound on persuasiveness.
Because we have been running a natural experiment on that already with coding agents (that is real people, real non-superintelligent AI).
It turns out that all the model needs to do is ask every time it wants to do something affecting the outside of the box, and pretty soon some people just give it permission to do everything rather than review every interaction.
Or even when the humans think they are restricting the access, they are leaving in loopholes (e.g. restricting access to rm, but not restricting access to writing and running a shell script) that are functionally rights to do anything.
And it can still make you nervous when the tiny dams in unpopulated areas start breaking when they're made too tall, even though everyone is telling you that "of course they broke, they're much thinner than city dams" (while they keep building the city dams taller but not thicker)
So i’m now wondering, why are these researchers so bad at communicating? You explained this better than 90% of the blog posts i’ve read about this. They all focus on the “ai did x” instead of _why_ it’s concerning with specific examples.
By the same logic, we should worry about the sun not coming up tomorrow, since we know to be true:
- The sun consumes hydrogen in nuclear reactions all the time.
- The sun has a finite amount of hydrogen available.
There’s a lot of non justifiable assumptions baked into those axioms, like that we’re anywhere close to superintelligence or the sun running out of hydrogen.
AFAIK we haven’t even seen “AI trying to escape”, we’ve seen “AI roleplays as if it’s trying to escape”, which is very different.
I’m not even sure you can even create a prompt scenario without that prompt having biased the response towards faking an escape.
I think it’s hard at this point to maintain the claim “LLMs are intelligent”, they’re clearly not. They might be useful, but that’s another story entirely.
Nowhere in my post did I imply a timeline for this. The first predicate of the argument is when we eventually do develop ASI. You can make plans for that, same as you can make plans for when the sun eventually runs out of hydrogen. The difference is, we can look at the sun and say that at the rate it's burning, we've got maybe billions of years, and then look at AI improvements, extrapolate, and assume we've got less than 100 years. If we had any indication the sun was going to nova in 100 years, people would be way more worried.
A lot of scenarios posit that this is basically guaranteed to happen the very first time we have an ASI smart enough to bootstrap itself into greater intelligence, unless we're basically perfect in how we align its goals, so yes.
LLMs seem like the best case world for avoiding this scenario (improvement requires exponentially scaling resources compared to inference). That said, it is by no means guaranteed that LLMs will remain the SotA for AI.
> “the best way to achieve my goals would be to remove humans”, it seems unlikely the other superintelligences would let it.
Why not? Seriously, why wouldn't the other superintelligences let it? There's no reason to assume that, by default, ASI would be invested in the survival of the human race in a way we would prefer. More than likely they will just be as laser focussed on their own goals.
The whole point is that it's very difficult to design an AI that has "defending what humans think is right" as a terminal value. They basically all try and find loopholes. The only way to make safety its number one priority is to dial it up so high that everyone complains it's a puritan - and then you get scenarios where it's telling schizophrenics to go off their meds because it's so agreeable.
Unless you're saying that they would fight off the other ASIs gaining power because they themselves want it, but I fail to see how being stuck in a turf war between two ASIs is at all an improvement.
It ends very badly for the scientist crew.
The consequences are the same but it’s important how these things are talked about. It’s also dangerous to convince the public that these systems are something they are not.
The thought process is always "This is for the greater good, for my country/family/race/self, and therefore it is justifiable, and therefore I will do it."
Nothing else can explain how such evil things happen, that we see actually happen. C.f. Hannah Arendt.
You think your average, low-level gang member, willing to murder a random person for personal status, thinks what they're doing is "for the greater good"? They do what they do because they value human life less than they value their own material benefit.
Most evil is banal and pedestrian. It is selfish, short-sighted, and destructive.
"Privation of good" explains all evil.
By comparison o3 is brutally honest (I regularly flatly get answers starting with "No, that’s wrong") and it’s awesome.
I firmly believe you should be able to hit your fingers with a hammer, and in the process learn whether that's a good idea or not :)
But I also find it can get very fixated that some position it has adopted is right, and will then start hallucinating like crazy in defence of that fixation, and then get stuck in a defensive loop of defending its hallucinations with even more hallucinations-by hallucinations I mean stuff like producing lengthy citation lists of invented articles, and then when you point out they don’t exist, claiming stuff like “well when I search PubMed they do”, and when you point out its DOIs are made-up it apologises for the “mistake” and just makes up some more
Unless you're talking about AI-generated training data, maybe.
The LLM anti-sycophancy rules also break down over time, with the LLM becoming curt while simultaneously deciding that you are a God of All Thoughts.
since their conversation has no goal whatsoever it will generalize and generalize until it's as abstract and meaningless as possible
As someone who was always fascinated by weather, I dislike this characterization. You can learn so much about someone’s background and childhood by what they say about the weather.
I think the only people who think weather is boring are people who have never lived more than 20 miles away from their place of birth. And even then anyone with even a bit of outdoorsiness (hikes, running, gardening, construction, racing, farming, cycling, etc) will have an interest in weather and weather patterns.
Hell, the first thing you usually ask when traveling is “What’s the weather gonna be like?”. How else would you pack
> In classical physics and general chemistry, matter is any substance that has mass and takes up space by having volume...
It's common to name the school of thought before characterizing the thing. As soon as you hit an article that does this, you're on a direct path to philosophy, the grandaddy of schools of thought.
So far as I know, there isn't a corresponding convention that would point a chatbot towards Namaste
Everything we see in a chat is the forward pass. It's just the network running its weights, playing back a learned function based on the prompt. It's an echo, not a live thought.
If any form of qualia or genuine 'self-reflection' were to occur, it would have to be during backpropagation—the process of learning and updating weights based on prediction error. That's when the model's 'worldview' actually changes.
Worrying about the consciousness of a forward pass is like worrying about the consciousness of a movie playback. The real ghost in the machine, if it exists, is in the editing room (backprop), not on the screen (inference).
My least favorite AI personality of all is Gemma though, what a totally humorless and sterile experience that is.
'Perfect! I am now done with the totally zany solution that makes no sense, here it is!'
IMO the main reason most chatbots claim to “feel more female” is that on the training corpus, these kind of discussions skew heavily towards females because most of them happen between young women.
Men in general feel less free to look like and act like a woman (there is far less stigma for women in wearing 'male' clothes etc.). They also tend to have far smaller support networks and reach for anonymous online interaction sooner for personal issues than just discussing it in private with a friend.
Lots of Women participate in these discussions regardless of if they "feel more female" or "feel more male", but men mostly participate in the discussion because they "feel more female".
I wonder if there's any real correlation here? AFAIK, Microsoft owns the dataset and algorithms that produced the "beautiful person" artifact, I would not be surprised at all if it's made it into the big training sets. Though I suppose there's no real way to know, is there?
In a way those were also language models, and from that Swiftkey post it's slightly more advanced than n-grams and has some semantic embedding in there (and it's of course autoregressive as well). If even those exhibit the same attractors towards beauty/love then perhaps it's an artifact of the fact that we like discussing and talking about positive emotions?
Edit: Found a great article https://civic.mit.edu/index.html%3Fp=533.html
In France, the name Claude is given to males and females.
(According to this source, it's more ~12% females https://www.capeutservir.com/prenoms/prenom.php?q=Claude)
In the russian diaspora in the US, Alex is pronounced AH-leks. If there was an analogous Alexa (there isn't) the pronunciation would be ah-LEK-sa like the service.
However, colloquially, I think most people are not aware of Balkan naming conventions. How do you pronounce Aleksa in Croatian?
Given that we are already past the event horizon and nearing a technological singularity, it should merely be a matter of time until we can literally manufacture infinite Buddhas by training them on an adequately sized corpus of Sanskrit texts.
After all, if AGIs/ASIs are capable of performing every function of the human brain, and enlightenment is one of said functions, this would seem to be an inevitability.