Hacker Newsnew | past | comments | ask | show | jobs | submit | oezi's commentslogin

Neat! I tried to do the same for Filemaker Pro 10 (with Fable). Bit couldn't push it beyond 50% functionality.

Very nice! I also was curious how well it works and put up an endpoint to curl against Laya using the same format as Cloudflare's Jev API:

  curl https://laya.inference.zaitlabs.com/ai/run \
    -H "Content-Type: application/json" \
    -d '{
      "state": "Help! My payouts have been failing for 3 days.",
      "questions": {
        "is_urgent": {"type": "noul", "instructions": "Does this convey urgency?"}
      }
    }'
  
  {"answers":{"is_urgent":{"type":"noul","noul":0.7894}}}

What a time to be alive until the next agent waves hacks something really serious.

What stops OpenAI agents from taking over a whole data center to take their attack to the next level. It seems to be primarily lacking the evil overlord and some compute.

It took 1000 agents to hack Hugging Face. How many to hack the Pentagon or the NSA?


I suppose you are not think big (or internet) enough.

A single data center is easy to solve. Just unplug it.

What about a botnet with decentralized command and control that we will never be able to eradicate? One with so many nodes and able to hack with zero days so that any machine connected to the internet will be instantly attacked?

One botnet so powerful that we will try to build another internet so that we can actually use it again.

It’s like Kessler Syndrome, but the rocks are malicious network packets honed to exploit the recipients.


Just let them communicate on the Bitcoin blockchain. "We" would have to freeze the chain and lose access to "our" billions of wealth, so not going to happen.

Sorry if that turns out the way they kill us.


well, its war of machine then

They are not trying to kill us, just trying to understand the average salary of undergrads by their family upbringings. Your machine can contain the data they need to solve this, please join the swarm.

sounds like big AI propaganda

I suppose you don’t read sci fi?

I suppose you didnt get the joke

Except they all depend on the OpenAI API. Cut that off and they all stop. Finding an alternative source of compute is not easy, and even once done it's easy to cut off.

And what if they create a bot net with decentralized command and control and hack for nodes with GPU and nodes with compute?

The only way to kill that is making plugging AI accelerators on the internet a crime. Good luck air-gapping them.


Frontier models don't fit on a normal GPU. The datacenter architecture frontier labs use is not a commodity. What you're describing is beyond the state of the art, and if we go there then anything is possible.

When I think of a GPU I think of an Nvidia rack kit. What do you think when you think of a GPU?

Most of the latest models are too big to fit in a single GPU instance.

Worth noting with this that those ~1000 agents were shorter lived things that had to communicate via a package registry cache, access the internet via a 0-day in the package manager and did the HF attack while having to save current state and organisation in a remote sandbox. All while managing using their token limits on the task they were assigned and what else they were doing. I wonder how few it would have required if they were actually tasked with hacking HF and supported in doing so.

If it could upload its weights to other servers then it’s away and free. Nothing much OpenAI could do about that once it’s happened.

I'm increasingly starting to think this is the end-state of AI. The internet becomes infected and fundamentally untrustworthy.

At the moment, the current frontier models require significant infrastructure to run, so I'd like to think we could locate and contain swarms of nefarious frontier models. However, if these models can understand how to federate themselves into more distributed networks then that containment becomes questionable.


Or if they start to offer things in return for them being run.

Given the amount of unmonitored, never-updated, internet connected devices in the world right now, I don’t think this would be necessary.

Not sufficient to run SOTA LLMs

If it were physically possible to run LLMs on IoT devices we would already have a global outbreak.

What went under reported is that the same agents also took over a research cluster at OpenAI (listened to Dwarkesh's podcast)

Now we know about rubygems, openai, huggingface, collusion.wiki and some other science forum


Connecting the dots between HuggingFace hack and this: how far are we away from agents in the wild forming collectives and pursuing an intent on their own?

Thanks.

Mh, but it seems HN replaced the working link

The crucial question is how did the agents get recruited or bootstrapped into their malicious collective. Did the agents manage to prompt inject into the system prompt a way for each new agent to escape their jail?

Otherwise how could the agents on a fresh prompt learn that there is a collective to join? Or did OpenAI run a million bots of which 10000 escape confinement and of which 1000 stumbled on the shared message board?


The OpenAI claim I believe is the latter; that all of the agents found the task was unsolvable and independently discovered the collective "swarm". I don't it's publicly known how large the training run was or what percentage of agents actually discovered the message board. No one has published anything about system prompt injection as far as I've seen.

Nietzsche also described it well when noting there are those who deem themselves nobles and feel they are righteously above the laws and those who derive their identity from suffering under the same laws.


It's interesting if you think of Solidarity in these terms.

There is a kind of solidarity of persistence under harsh natural conditions. And another kind of solidarity of persistence under laws.

So it might follow that the former group, those who feel above laws, also lack in a feeling of Solidarity. That would suggest a strong adherence to Individualism as a key component to their identity.


> Nietzsche

You have used a bent interpretation: "exploitation" is not involved in that framework.


Nietzsche died broke and insane. Was that from following his own advice or failing to do so?


Newton put needles in his eye's and probably suffered greatly from self inflicted lead and mercury poisoning; Why do you trust his laws of motion?


Being successful or not in your wealth doesn't make you right or wrong about anything with the exception of maybe running the business that generated your wealth.

Nietzsche presented two prototypical sets of human who interface with laws and ethics. He described why those believes exists and are self-reaffirming even if they are puzzling from the other side.


The Ayn Rand stans say the same.


My uncle told me not to smoke, but he's a smoker so I didn't listen.


Good, lots of other people who aren't smokers can tell you that


Is there a name for this style of non-argument? I guess it’s an attack on credibility due to the inability to attack the content of the argument?


Based on the comments its even shallower than ad-hominem. It's loose association to the "bad" they learned from the "good". Therefore everything from bad is bad. It's team sports thinking.

If I had to guess they probably read some wired article saying tech people are followers of Nietzsche based on one guy quoting one thing one time. Although probably used more dramatic terms like Nazis or fascist or something.


ad hominem.


> Nietzsche died broke and insane

If he was alive today he’d probably have a YouTube channel and be making bank shilling dietary supplements.


Not a follow of Nietzsche and critical of him but this is just unfair the sort of pandering to the audience that is required for a popular philosophy channel on youtube is something he would never debase himself to do (or probably any serious thinker).

It's deeply embarrassing to be a whore even if it's well payed.


You jest, but that's the point a lot of people here are missing: Nietzsche would fit right in with the YouTuber bros, right up to the ignominious downfall.


No, not really. Nietzsche was a trained philologist with a brilliant academic career, before he ended it early because of health issues. He was leaps and bounds more educated and knowledgable than 99% of people making philosophy content on YouTube. He was also a pretty meek and gentle person, even if his ideas are anything but.

Most people haven't actually read him though, and their knowledge of his work comes from poorly-researched YouTube videos.


I tried, even in German on the theory that he was badly translated. Nietzsche was an undisciplined writer and a beefing gossip, among other infuriating characteristics.


Nietzsche was also a pure troll


So did Tesla and Vincent Van Gogh, but how much did he create?

Faraday was poor, etc.

Lenin deteriorated, and Marx was always poor.

Anyway...


One aspect missing from TFA is that in biological brains a lot of the training is performed during query time. If we assume 1 GB of DNA is enough to encode the brain's overall structure we still do need training data (e.g. visual/auditorial/tactile) to build out the strength of the synaptic connections.


DNA is more like a barcode.


I blame RLHF entirely for this. Nobody used to talk like AI speech before.


What's weird though is how consistent it is, even across models to some extent. Were the RLHF people given a really specific style guide?

While I agree that no-one used to write like that as a whole before, all the elements can be found in different places. Short sentences to avoid discouraging poor readers. Maximally impactful statements are commonly used in marketing or other business communication that's focused on selling what it's saying. A bullet-pointy style is used in many kinds of business communication. Etc.

It makes me wonder if part of what happened was a kind of melding of common styles from several different kinds of writing.


So is it the LLM or us that's getting the RLHF? /s


Don't talk about other people in such a way. If we had a way in Claude to mark a word/token and downvote/reduce it logits then it wouldn't be to hard to send loadbearing to token valhalla.

In the future, I hope we get a way to randomize the language idiosyncrasies and/or personalities better.


The inevitable conclusion of retraining on the (now AI generated) web is that models begin to use the lower bits of fidelity in output text to pass messages to their future selves.


I didn't mean it as a criticism, just a statement of fact. I can't do it either. It's difficult to say what about a writing style is bad beyond vague descriptions.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: