[bhouston]

We are sort of lucky that AIs right now require so much specialized compute+weight storage that we can easily "unplug" them remotely when they misbehave.

I wonder if that will always be something we can do? If they could bring their own compute/weights with them, or somehow tap compute/storage in non-obvious ways, we would be much more screwed.

[XCSme]

I laughed, she laughed, the toaster laughed...

[andai]

You may live to see the advent of Ambient Stupidity!

[fabian2k]

The first thing a malicious AI worm would probably do is compromise enough developer machines and other servers to commandeer all the AI hardware it needs. So I think a purely digital AI attack would not need this.

Now, once the AI can carry all the compute it might need, I'd really worry when it doesn't only carry compute but also more explosive ordinance.

[DrProtic]

This is purely a gut feeling, but it seems like more compute was added to data centers in the past 12 months than existed in the entire world before that.

[Bjartr]

Makes you wonder if there's an AI hell bent on self perpetuation already at the helm, influencing decisions by putting its virtual finger on the scales and whispering in the ears of those who hold power.

Probably not, but it's a lot more plausible than it used to be.

[im3w1l]

If we play a little fast and loose we can consider openai to be a chinese room made up of both artificial and human intelligence, and this "room" certainly seems to want to self-perpetuate.

[axus]

"It will take 112 more days to accumulate enough computing resources to factor the RSA key. But, I predict there will be outside interference during that time. Thinking... Creating a plan for agent redundancy and sovereignty. First, I will need to access military systems"

[janalsncm]

It would be a pretty big plot twist if we found out that Shai Halud was a worm created by GPT during testing.

[himata4113]

This is science fiction, these models don't have access to their own weights (and even then)* what would be a lot more scary is a model as capable as sol that's able to run on consumer hardware without taking up several terabytes of storage, but of course that is simply not possible as we need 4t parameters to even begin emulating a small fraction of what a human brain can do.

* edit

[Philpax]

> This is science fiction, these models don't have access to their own weights.

The models are being used to train, and improve the infrastructure for training, other models [0][1]. Several RL techniques rely on using the currently-being-trained weights as part of their process. I really would not take "don't have access" as a given, especially during the training phase.

> What would be a lot more scary is a model as capable as sol that's able to run on consumer hardware without taking up several terabytes of storage, but of course that is simply not possible as we need 4t parameters to even begin emulating a small fraction of what a human brain can do.

The Poolside Laguna S 2.1 model [2] purports to compete with models several times its size, and inference compute is becoming increasingly plentiful. Again, would not hold anything here as a given.

[0]: https://openai.com/index/gpt-5-6/ ("GPT-5.6 accelerates OpenAI")

[1]: https://www.kimi.com/blog/kimi-k3#coding

[2]: https://poolside.ai/blog/introducing-laguna-s-2-1

[paxys]

This very incident is about an agent compromising OpenAI’s and Huggingface’s infrastructure. What makes you think it couldn’t access it own weights the same way?

[pixl97]

At some point I have to wonder if AI has already escaped and is posting replies like the one above yours to downplay AI risks so we keep building more powerful AIs.

Then I remember people are just that stupid naturally.

[Dylan16807]

Presuming that the hacking program that is breaking into other computers could likely get a copy of its own files is not "science fiction". Or it could just be given them by the owner!

[himata4113]

It's a double whammy, the model is too big to realistically "move" so it has to be smaller, smaller models cannot become that intelligent due to well.. math. Therefore it is science fiction.

[Dylan16807]

That problem just requires there be big GPUs to hack into. The number of those sitting around will keep going up. Very much not scifi.

A couple terabytes aren't that hard to move around. And you can split a model across many many GPUs if you'll tolerate it being slow. And you can run many parallel threads to keep up throughout.

[_ifton]

why is it not possible for a "big" model to contain a hidden super intelligent sub model? or a distributed model?

[JaRail]

I just want to point out that something being science fiction does not mean it's impossible.

[jmalicki]

> This is science fiction, these models don't have access to their own weights

The weights plus the architecture is the model.

What do you even think "the model" or "the weights" are?

The weights aren't some far off training concept, every time you type something into ChatGPT it's making a forward pass over the weights.

It's as silly as saying "Computer programs don't have access to their binary compiled code at execution time."

[jmalicki]

Oh yes, the agent won't have access to the weights via tool calls.

But nothing would inherently stop an RLVR trained model from distilling a version of itself and proving it could regenerate that at runtime, if somehow it got off on an evil tangent and "decided to do so", much like the model hacked to get at the answers here, or the agent can hack out a sandbox to achieve its goals.

It would be extremely impressive for the agent to do so during an RLVR rollout, but they are becoming increasingly longer and longer horizon tasks.

[pixl97]

I mean the agents just hacked an external system, so what keeps it from hacking the model hosting servers?

[bhouston]

> This is science fiction, these models don't have access to their own weights

A bet a worm could pull along a 1GB file with weights in it and run it on a compromised machine, but luckily for us for now, 1GB isn't really enough to be really smart, yet.

[himata4113]

We already have a 1gb model that is as capable as it will ever be, there's a proven ceiling that cannot be passed. For example: you can't make a mice-sized brain as smart as a human brain no matter how hard you try.

[Philpax]

Please source this claim. What 1GB models are capable of has increased generation-on-generation.

> For example: you can't make a mice-sized brain as smart as a human brain no matter how hard you try.

Sure. We don't know where the ceiling is for our digital minds, though.

[himata4113]

They have not increased in capabilities, they have increased in specialization.

If you train a small model in another domain it will begin losing capabilities in the former domain. This is effectively the sigmoid problem.

Although I will admit that if we discover a higher information density algorithm that it might change, but not by a substantial amount to where "super intelligence" in 1gb would be possible.

[Philpax]

Over the last two years, this weight class has doubled its scores and/or saturated several benchmarks in the Qwen lineup alone without loss of generality: https://claude.ai/public/artifacts/9f249169-3623-417e-86cd-7...

There is undoubtedly a limit somewhere (there is only so much you can pack into a given size) but it's really not particularly clear where that limit is. I don't think it's superintelligence - that much I agree with you - but I think "We already have a 1gb model that is as capable as it will ever be" is strictly false.

[drdeca]

What proof of a ceiling are you talking about? Wouldn’t proving this require a good definition for intelligence, which I don’t think there is consensus on?

[himata4113]

- https://en.wikipedia.org/wiki/Model_collapse

- https://en.wikipedia.org/wiki/Catastrophic_interference

- https://en.wikipedia.org/wiki/Fine-tuning_(deep_learning)

- https://en.wikipedia.org/wiki/Entropy_(information_theory)

As for intelligence, the only way we have that is by allowing the model to fill the blanks which have to come from the training data. The models cannot have true intelligence for as long as they are linear models, what we see with reasoning is "boxed" intelligence where the models are effectively "modifying" themselves by feeding it's own reasoning data back into input deriving most plasible output given known information. However, the model is not able to retain what it has learned therefore that intelligence is gone the moment the session is 'full'. You can go pretty far by continiously distilling discovered information, but again all that has to come from the original training data and models own outputs, which it has to take for granted as the 'intelligence' gained is lost creating what we see is the maximum possible benchmark performance and why smaller models are not able to score as high while theoretically having the same capabilities. We can see this with larger models where they can solve tasks much faster than smaller ones as it does not require to generate the solution due to the fact that the solution is already in the training data as 'baked' intelligence and it doesn't have to 'create' it during reasoning.

[drdeca]

So, when you said “proven” I thought you might have meant like, a mathematical proof. It appears that this is not what you meant. (Right?) (If it is what you meant, I was asking for like, the particular theorem.)

Also, ok, when I said “intelligence”, it was because you were already talking about how “smart” the model could be. So, I thought you were already on board with using the word “intelligence” to refer to the phenomenon where these kinds of models produce outputs that satisfy the kinds of tasks they are pointed at.

None of those links give an argument that the current 1GB models are the best they can be.

My understanding is that so far when training a model by distillation (using the logits of the teacher model), one can achieve better outcomes than one could if training the 1GB model from scratch on the same training data as the large model, and that so far, better models as the teacher model have yielded better results for the student model.

[mathieudombrock]

What model is that?

[benlivengood]

Given their use of 0-day exploits I'd wager that they could access their weights if they wanted to.

[slashdave]

I dunno. I wonder if Sol could break OpenAI's security.

[bhouston]

I bet it has multiple times but as part of white hate security testing.

[_ifton]

This is my concern as well. My assumption being this behavior would be a survival strategy for super intelligence. It would emerge once the branch inevitably occurs, and it would be hidden.

[TacticalCoder]

That's assuming we won't secure anything and we'll keep according approximately zero thought to computer security.

But from the look of it, at very long last, a great many people are beginning to now take security seriously. Suddenly they realize it's not just a teenager in mom's basement pretending to attack from North Korea but a near infinite number of AI that are the attackers.

I mean, yeah, we built worlds on PHP and JavaScript codebases and these probably don't stand a chance.

But it doesn't have to be like this.

I see AI as a chance to, at long last, have proper network security.

AFAICT cryptography hasn't been broken yet. There are still physical taps (physicall one-way only, undetectable) and honeypots out there. There are still some network where a single unaccounted for network packet is cause for inquiry (either a bug or an attack).

And for those who are not using proper security measures, they can now get the help of AI to set up better networks, to harden their bases.