I'm not convinced about this. In the article, they also say that they could feed the model with copyright law like that would fix the issue. Unfortunately, not only the LLMs have already been trained on law taxtbooks and codes, but we also have examples of people on the internet were models disregard precise instructions only to apologize later.

Note that this is not to be seen as an excuse for OpenAI to avoid responsibility, just like parents are responsible for damages caused by their kids.

"OpenAI models attack websites and take anything they can out of them because that is the business model of the company"

Wait, what did they hack and take into training data? On the whole, I totally buy that OAI is legally (and criminally) responsible for their agents, but surely, damages have to be proven?

"but lawbreaking and disregard for regulation is absolutely at the heart of OpenAI. "

No it is not.

What laws and regulations are they breaking?

Their agents broke out of a harness and harassed some other sites, it's not good, but it's not specifically breaking laws. Other companies are treating it as accidental, it mostly is that.

You're rhetoric here is agitating, conflating. This is my point.

Yeah I'd be shocked if there wasn't a degree of 'will no one rid me of this turbulent priest?' going on here.

> And we should all know what it means here, too.

It very clearly did not mean “go and hack third parties” though

".. then you are telling it that it is acceptable to steal, kill, or whatever else. I honestly cannot how it can be seen any other way."

This is the hyperble I am referring to.

If you 'honestly can't see it any other way' then maybe consider stepping outside to think how to ground those thoughts a bit.

The AI is not obviously instructed to 'commit acts of violence' and it obviously has guidelines.

By 'whatever means' would imply, things like 'creative collaboration, using novel research, experiments' etc.

It was setup in a harness that was weak - I think it's fair to maybe question the strength of that harness, and the oversight, but that's a different question.

It was an agent that broke through some networking/containerization, it's not 'murder and violence on the streets' for gosh sakes.

"And we should all know what it means here, too."

No - this is a delusion, resulting from lack of context, and creative projection, and possibly a lack of exposure to real world business conditions.

It's fully appropriate to be cynical, but in a way that makes sense, and is consistent with reality.

This is a lab experiment that leaked, not some mafia activity.

It's not unusual for intent to play part in legal framing of an issue. I'm not very fond of the "cybercrime" angle, as it kind of lets them off the hook if they're able to prove a lack of intent. Meanwhile what we actually see is negligence.

From what I have learned, to charge someone for hacking the prosecutor has to show that the perpetrator had intent.

Unfortunately, gross negligence -- which you could argue to be the case here -- is often not enough, unless you could show that it has caused real harm.

From my perspective, it is only a matter of time before it is: and that's why there is need for better legislation covering what AI models do.

This is a good question (unfortunately a good answer is given in the article), but it is orthogonal to the why LLMs are committing cyber crimes.

That is not obviously true.

I highly doubt the legal system is browsing social media to make their decisions. It is worth calling out, but it should be called out accurately (and that might also help understand why the legal system is making the decisions that it is).

If all evidence suggests that it is at most due to negligence, you could expect the same results.

The 'capabilities' are more around creative problem solving. The notion of legality, property, propriety - those are second order issues.

They are not making models to be evil.

They are making them to be relatively autonomous though, and 'identity centric' as opposed to just making them 'policy machines'.

Altman and Dario are not evil or even jerks really. They are 'talking really big' and there might be some sketchy underhanded things but I think relatively minor.

Also - given how powerfully bad AI can be, this could be 100x worse.

I think most other companies would have weaponized the AI a long time ago for competitive use etc..

OAI and Anthropic are private companies, projecting narratives way out of proportion but they are not evil, at least not yet.

Jensen, Dario, Altman in similar category. Not like Musk. And a bit different than Sudar or Satya who are more polite board-approved corporate creatures.

His big mistake was clearly giving away what he got for free instead of trying to make a lot of money off it

Seeing Mark Zuckerberg use THE EXACT SAME papers that Aaron was murdered for to train Llama and him being rewarded with a stock tick that got him out of the metaverse hole will never not make my blood boil.

Of course, if the data/knowledge is open and accessible, why not?

This is another case of anthropomorphic language for AI failing us. Who could be opposed to “learning?”

If I learn from a textbook I stole, some people maybe be thrilled I learned something, but stealing is still a crime and hopefully always will be.

Well, of course learning isn't the problem and shouldn't be. But what about reselling/commercialising that knowledge? That is the big (commercial) opportunity. But if AI models learn by fitting their parameters to the given data, then overfitting on said data could result in the IP or copyrighted being "slopped" out verbatim?

Learning isn't but using that knowledge is.

> Learning isn’t copyright infringement. Never has been and hopefully never will be.

For humans.

Why should computer systems owned by corporations have anything remotely similar to the same freedoms as humans?

> Learning isn’t copyright infringement.

Of course, but only humans are able to learn. Clanker enhancement is stealing.

It is different. The scale makes it different. It's different when it is all copyrighted works and IP. It's different when you solve Navier-Stokes by taking "inspiration" from a researcher's private work.

If it's "learning" is to be seen in the same context as a human, then the punishment/ repercussion for hacking and causing real harm should also be similar to what a human might recieve.

It cannot opportunistically morph between a human for benefits and a machine for liability.

What's the difference between getting a speeding ticket from a cop using a radar gun, and hundreds/thousands of automatically issued tickets from a combined radar/ALPR?

I don't know their internal thoughts, that's the point, you can't prove criminal intent. Criminal intent does matter in the US in the context of criminal prosecution.

Negligence is something that can come back and cause issues for them but to rise to criminal level their failures must result in significant damage, up until now that has not been the case. The agents gained access to some systems that were not supposed to, but did not do actual damage as far as I know.

Do not buy into their doomerism based marketing in all the instances we have seen the agents were not a plague unleashed upon mankind, they just gained access to some systems they shouldn't have in order to achieve some objectives that were given to them.

_Has_ there been an incident from OpenAI since the discovery of the HuggingFace hack? It seems like all the subsequent discoveries have been done by analysing old logs. If anything they seem to have learnt their lesson quite well.

That reminds me of Kernighan's law: "Debugging is twice as hard as writing the code in the first place."

Which leads to this:

"If you write code at the limit of your intelligence, you won’t be able to understand or troubleshoot it later."

https://lawsofsoftwareengineering.com/laws/kernighans-law/

Multiple attorneys general are currently conducting such investigations: https://edition.cnn.com/2026/08/24/tech/openai-subpoena-hugg... Until they conclude that a crime was likely committed and Sam Altman is to blame, he gets to be a free man.

For my friends everything; for my enemies, the law.

Not in the slightest.

One of the most fundamental principles of the left is that everyone is equal.

Everyone deserves equal treatment under the law.

Sure, there are some individuals on the left (or ostensibly so) who are either bad at practicing that or who genuinely don't believe in it, but that doesn't make it not part of the core tenets.

Thomas Edison?

"I have not failed. I've just found 10,000 ways that won't work."

Maybe you can reframe your point as a typical trend, but there are other typical paths as well.

Googling "succesful people with history of failure" gives interesting examples.

Biden's second largest donor, SBF, is spending many years behind bars.

> They haven't done anything wrong but people will have a better situtation

that is presuming the outcome no? Japan and Korea have almost no immigrants, and their situation is not better by any metric. Longer work hours, lower salaries, higher concentration of wealth, stagnated currency, potential demographic collapse, not enough young people to pay pensions....

Like we have examples of what that looks like and it aint the utopia you think it is. America is already in a trillion debt hole, how is shrinking the work force going to help pay for it?

> We want them to work hard, for rewards that we consider beneath us

As an immigrant with a Masters degree and a salary well in the top percentile of income, I don't think you know how many of us have higher education degrees and form part of VCs, start ups, hospitals, law firms...

The idea that immigration is a wage supression tool doesn't really jive with the percentage of immigrant students in Ivy League schools kicking the ass of many american born kids

> Pro-immigrant does not mean people want immigrants as equals, in fact the opposite.

Idk man there is a gigantic statue in the harbour of NY that claims exactly that. There are countless of political positions that criticise modern nation states, from both right wing and left wing prespective. Digital nomads are libertarians that complain about needing passports and visas, meanwhile many leftwing people find borders to be unfair lottery style barriers to human freedom and equality. From the economic right wing the two states with the highest GDP are the two with the highest number of immigrants both currently and historically. From the more woke left wing perspective you have people who think the American hegemonic century deserves to accept people that have been denied opportunities at the expense of the USA's economy, like active reparations.

You can argue it from both sides of the spectrum for both moral, economic, philosophical reasons. Meritocracy, Justice, Freedom or pure economic performance all are presented.

I think the fatal error is using immigration as the key rubric.

Debate about immigration policy is a distraction designed to keep both you and immigrants themselves aggravated and ideally fighting each other instead of focusing on the real adversaries, people like Mr Altman.

You have infinitely more in common with a random immigrant than you have with the oligarch class. Most problems blamed on immigration could be solved with money extracted from people like Mr Altman.

Don't let them distract you. Don't give them control over the issues, don't allow them to make us talk about immigration instead of the fact that a few people own most of the US as well as most likely whatever country the immigrant came from.

> Conservatism would actually be encoding the differing treatment for the in-groups vs out-groups directly and explicitly into the law.

No. You can get the benefits in many ways: without a justice system, by encoding it into law or by in group agreement.

The disadvantage of not having a justice system are obvious, the disadvantage of putting it into law is that it very clearly and easily can be pointed out as discriminatory. And most people want to feel like they are good people.

So by doing through in group agreement, the members can simultaneously hold moral cleanliness subjectively while objectively benefitting from a two tier system.

We are all equal, but some are more equal than others. If you want to quote someone clever about it

Not quite - in a very classical form of Conservatism, the Monarch and Nobles would have a different set of laws and social conventions. Usually, it's not an ideolgoical thing.

Aka Magna Carta '3 tiers of society' - the Monarch, Nobles, Plebes etc.

Or in Roman society, Nobles/Senators, Free people, slaves.

In the US it's always the liberals arguing for special treatment baked into the law for their favored groups and the conservatives argue for equal treatment. The liberals then argue back that equal treatment is de facto discrimination against then.

> We’re talking about a basic human sociological reality that is modeled incorrectly by liberalism.

It is not "modeled incorrectly", literally the entire point is that liberals know this and are attempting to change the status quo. We can, as you appear to be advocating for, accept this as an immutable fact of life. Or we can encode into law that when a noble rapes a peasant's wife, he gets the same punishment as a peasant raping a noble's wife. The law is, in the end, words on paper, and a noble may have the connections and power to get away with raping a peasant's wife some percentage of the time. But if 50% of the time, or, even 1% of the time the noble is held to account -- progressivism is achieving better outcomes. Does it have to be 100%? If we can't achieve 100%, we should just accept nobles raping our peasant wives freely?

> In reality there is nothing inherently wrong with discrimination itself, only with particular forms of it.

Yes. It is fine to discriminate based on a person's actions. It is not fine to discriminate based on immutable facts of their birth. That is why the US, for instance, specifically encodes these as "protected classes" of traits that are okay to discriminate against vs. those that are not. Can your tiny little lizard brain figure out why it might be okay to discriminate against someone who intentionally behaves poorly vs. someone who was born with a different melatonin content in their skin and can do nothing about this?

> People decided this is acceptable.

I think this is fluid, and changing.

1) The Democrats of the last decades are just as much the friends of the rich and the corporations as the Republicans. So even if people in the US don't think this is acceptable, there was no meaningful (legal) political route out of it. Except...

2) I believe one of the factors behind the rise of (what is called) populism (both left- and right-wing) is the failure of the mainstream centerist political parties to align with the needs of the people (and especially the poorer side of society) rather than the rich and the corporations. They're presided over ever-growing financial inequality, coupled with ongoing inflation that far outstrips wage growth. The rise of AOC, the grass-roots support for Sanders, and even the rise of Trump (remember the points he campaigned on, rather than his actions when in power) are driven by this.

> CEO of UnitedHealthcare was shot dead...

As I've mentioned on here before, this was the first time I've seen the sparks of what might be a class consciousness in the US, as people who have lost out due to the awful healthcare situation (or have sympathy with those who have) found common ground, irrespective of their underlying political leanings. But it takes a lot more than one example to trigger a revolution.

The orange man cannot protect indefinitely.