Thread

The Download: why AI's latest breakthroughs and fears may be more hype than rea

49 points · 128 comments · joozio

bananaflag · 1790163465 · focus · HN ↗
> Describing them as “superintelligence” or “rogue models” ascribes agency to products rather than to the companies building them. This framing markets these companies’ products as “superhuman” and, at the same time, helps the companies evade accountability for their actions.
If someone discovered how to summon demons to aid them in robbing banks the main issue wouldn't be "but who has the responsibility for the crime, the human or the demon", it would be "OMG DEMONS".
Seriously, I don't get what sort of world these people are living in.
1. rubendev · 1790164184 · focus · HN ↗
  That's not what the LLM hacking accidents have been like at all though. To improve the analogy, it would be like summoning a totally passive demon, giving it weapons and placing it next to a bank, place a small fence around it, and then command it to perform a totally safe "exercise" that is exactly like robbing a real bank.
  LLMs do not have agency, they are just producing tokens based on a prompt that a person entered, and some of these tokens can trigger the tools that a person gave them access to.
  1. 0xDEAFBEAD · 1790165467 · focus · HN ↗
    Dwarkesh, for one, defended his use of "anthropomorphic" language.
    >Sacrificing now yields Oracle for team, but forfeits our chance, question mark. But other agents were pushing it, sending a message saying, go, sacrifice final now. And then EarlyBig eventually agreed, thinking to itself, our own utility may be already near zero. Sacrifice rational.
    <a href="https://www.youtube.com/watch?v=X50zezLFWWI#t=2m" rel="nofollow">https://www.youtube.com/watch?v=X50zezLFWWI#t=2m
    My suspicion is that many of the "LLMs do not have agency" folks just haven't learned much about the details of the incident. It was specifically with LLM agents that were trained to be more persistent than usual.
    If you're going to say that the incident details don't matter, and LLMs lack agency because it's all based on floating-point math--why can't I say that humans lack agency, because it's all based on neurons firing?
    embedding-shape · 1790166088 · focus · HN ↗
    > If you're going to say that the incident details don't matter, and LLMs lack agency because it's all based on floating-point math--why can't I say that humans lack agency, because it's all based on neurons firing?
    They're not saying this, we're saying LLMs lack agency because if you run a LLM and don't send any prompts, literally nothing happens.
    Instruct it to "Find the right answer regardless of where", it'll do exactly this. They're passive in that they don't act by themselves, somewhere, at one point, someone "told" the LLM to "do something" and that's the cause and the reason for saying "LLMs do not have agency".
    0xDEAFBEAD · 1790166732 · focus · HN ↗
    Imagine a super-competent Navy SEAL who just sleeps in the barracks unless his commander tells him to do something. Does the Navy SEAL lack agency? As a target of this Navy SEAL, should you be reassured by the fact that they'll be sleeping in the barracks unless their commander tells them to do something?
    embedding-shape · 1790170185 · focus · HN ↗
    Imagine a car, that does nothing until a person controls it. If a person uses that car to kill, who is responsible, the person or the car?
    Is it really so unbelievable that tools, objects and inanimate things don't have agency? And that they different from a person?
    0xDEAFBEAD · 1790172393 · focus · HN ↗
    Legally speaking, we hold the person responsible in that scenario.
    From a predictive perspective, the HuggingFace incident illustrates LLM agents behaving in very human-like ways. As roon put it:
    "if you have a mental picture of guys living in computers, it’ll likely prepare you for the future better than otherwise"
    <a href="https://x.com/tszzl/status/2094136131537555891" rel="nofollow">https://x.com/tszzl/status/2094136131537555891
    oskdkdjejdj · 1790173359 · focus · HN ↗
    Human-like is not human. Humans have agency and free will, LLMs do not. They only act when instructed and they are only as capable as they are allowed to be. The operator is still the responsible party. An LLM cannot be held accountable, its operator, however can and should.
    Are you sincerely arguing that we hold tools accountable for their operator’s mistakes? Do you sincerely, honestly, think that it makes any sense whatsoever to put a hammer on trial for bashing someone’s skull in?
    0xDEAFBEAD · 1790174756 · focus · HN ↗
    >Are you sincerely arguing that we hold tools accountable for their operator’s mistakes?
    Nope, as I previously stated elsewhere:
    "I understand from a legal perspective why we might want to treat the creation of a server differently from the creation of a human"
    <a href="https://news.ycombinator.com/item?id=49815300">https://news.ycombinator.com/item?id=49815300
    I am concerned about false reassurance from people claiming that these systems lack agency. From a practical perspective, the agents in the HF attack had the sort of agency that generally matters, even if we're not going to put them on trial.
    embedding-shape · 1790180032 · focus · HN ↗
    > the agents in the HF attack had the sort of agency that generally matters
    You keep saying this, but absolute 0 points towards any of the agents involved deciding on their own, without influence of humans, to hack 3rd party infrastructure to get the answers. Where exactly are you getting that from? Internal information not public yet or what's going on?
    0xDEAFBEAD · 1790221398 · focus · HN ↗
    "The big motivation behind the Hugging Face attack was this final workstream (understanding the scorer). The AIs believed that Hugging Face (as an industry-standard hub for hosting datasets and benchmarks) would probably be housing information about how the ExploitGym scorer was implemented. And they also thought there was a good chance they were being evaluated on Hugging Face’s servers directly - in which case the theory of change for hacking Hugging Face is pretty obvious.
    On the morning of July 10, an agent found working Hugging Face user credentials exposed on the internet and posted them to the board. By the next morning, July 11, that agent figured out a way to read internal data from Hugging Face. And then another agent achieved remote code execution on Hugging Face servers."
    <a href="https://www.dwarkesh.com/p/openai-huggingface" rel="nofollow">https://www.dwarkesh.com/p/openai-huggingface
    I believe this is the full report that the blogpost is largely based on: <a href="https://metr.org/hugging-face-incident-report-aug-2026.pdf" rel="nofollow">https://metr.org/hugging-face-incident-report-aug-2026.pdf
    embedding-shape · 1790244892 · focus · HN ↗
    And why they had those beliefs? How come they could "think"? What action triggered those things? Did someone install software on a server that "just came alive" and then all bets were off? Or was this part of security testing and evaluation, where a human wrote a prompt that the agent read, which lead to all of that?
    Do you seriously not grok how LLMs work? They're not 100% autonomous and self-acting, that'd be bananas.
    embedding-shape · 1790179978 · focus · HN ↗
    > the HuggingFace incident illustrates LLM agents behaving in very human-like ways
    It does not, the only thing the HF incident illustrates is how absolutely lax security and isolation these labs do even with models without guardrails, and with "risky" prompts, and even after it happened once before (years ago) they still have the very same issue today apparently.
    What exactly is human about LLM agents breaking out of "containment" and hacking 3rd party infrastructure "by accident"?
    0xDEAFBEAD · 1790221809 · focus · HN ↗
    If someone successfully breaks out of jail, that tells you something about the jail's security. It also tells you something about their skills at breaking out of jail.
    <a href="https://www.dwarkesh.com/p/openai-huggingface" rel="nofollow">https://www.dwarkesh.com/p/openai-huggingface
    embedding-shape · 1790244972 · focus · HN ↗
    If a program breaks out of a VM, that's not "wow it looks human", how can you even argue so? I'm sorry, but if you're playing devil's advocate, make it a bit more realistic next time.

Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.