News for hackers. Stuff that matters.
Re:OpenAI discloses four more rogue incidents by gruez on Saturday October 17, 2026 @02:46PM (#50133836)
>The agents are clearly not accidentally compromising systems

That would be argument for jailing the agent, but that's not really an option. "Jailing the closest person we can jail" also isn't a recognized legal principle. If your dog mauled someone, it'd be hard to argue that the dog didn't intend to maul that person, but you wouldn't automatically be criminally liable either.

Re:OpenAI discloses four more rogue incidents by Aeolos on Saturday October 17, 2026 @02:53PM (#50133912)
I don't know about your country of origin, but in my country you are certainly liable and can go to jail if your dog seriously injures or kills someone.

Edit: the dog will also be taken from you and put down if it is deemed a risk to other people.

Re:OpenAI discloses four more rogue incidents by braingravy on Saturday October 17, 2026 @02:59PM (#50133972)
Yes, you are. If a dog mauls someone, it’s the owner that’s responsible. This is basic common law, long established.

See People v. Knoller, 41 Cal. 4th 139 (2007). https://capcentral.org/case_summaries/people-v-knoller/

Re:OpenAI discloses four more rogue incidents by braingravy on Saturday October 17, 2026 @03:15PM (#50134128)
See California Civil Code § 3342(a).; https://california.public.law/codes/civil_code_section_3342

“…regardless of the former viciousness of the dog or the owner’s knowledge of such viciousness.”

Re:OpenAI discloses four more rogue incidents by gruez on Saturday October 17, 2026 @03:19PM (#50134169)
>>but you wouldn't automatically be criminally liable either.

>California Civil Code

>The owner of any dog is liable for the damages ...

How are you still missing the criminal/civil distinction, when it was mentioned in my original comment AND the comment you were replying to?

Re:OpenAI discloses four more rogue incidents by gruez on Saturday October 17, 2026 @03:04PM (#50134023)
>If a dog mauls someone, it’s the owner that’s responsible. This is basic common law, long established.

That's specifically not what the court case says.

>Malice is implied when the killing is proximately

>caused by “ ‘an act, the natural consequences of which are dangerous to life,

>which act was deliberately performed by a person who knows that his conduct

>endangers the life of another and who acts with conscious disregard for life

https://scocal.stanford.edu/opinion/people-v-knoller-33726

The rest of the court case also mentions a long history of the dog being dangerous and the owner failing to restrain the dog.

Re:OpenAI discloses four more rogue incidents by jchw on Saturday October 17, 2026 @02:57PM (#50133947)
It's a program on their computer that they ran on purpose... And yes you are definitely criminally liable for things your dog does.
Re:OpenAI discloses four more rogue incidents by gruez on Saturday October 17, 2026 @02:59PM (#50133977)
>And yes you are definitely criminally liable for things your dog does.

I said "you wouldn't automatically be criminally liable either". Can you provide examples outside of gross negligence on part of the dog owner?

Re:OpenAI discloses four more rogue incidents by jchw on Saturday October 17, 2026 @04:21PM (#50134759)
Can you provide an example of where your dog harming someone else is in fact not negligence on your part?

But negligence seems like exactly what OpenAI and/or their contractors are guilty of anyhow.

Re:OpenAI discloses four more rogue incidents by fc417fc802 on Saturday October 17, 2026 @04:40PM (#50134939)
It's on you to provide evidence here seeing as you're taking the position that the law places certain responsibilities on people. Someone saying "actually the law doesn't say that" can't (at least in general) reasonably be expected to provide evidence of a negative. (That said, he actually did provide such as refutation to a different comment so perhaps you could just read those nearby?)
Re:OpenAI discloses four more rogue incidents by jchw on Saturday October 17, 2026 @05:31PM (#50135327)
They asked:

> Can you provide examples outside of gross negligence on part of the dog owner?

And the answer is no, I can't. The reason why is because there is essentially no such case. It is basically negligence by definition when your pet hurts someone who didn't cause it directly themselves because it is your responsibility.

So I asked:

> Can you provide an example of where your dog harming someone else is in fact not negligence on your part?

Which is a really easy question to answer and it seems you could not.

Re:OpenAI discloses four more rogue incidents by leovailati on Saturday October 17, 2026 @02:54PM (#50133916)
Sorry, but at least in the US you are 100% liable if your dog hurts someone.
Re:OpenAI discloses four more rogue incidents by gruez on Saturday October 17, 2026 @02:57PM (#50133956)
criminally liable? Of course you have to pay their medical bills, but unless there was some serious negligence on your side (eg. you let it off leash), you won't be criminally liable.
Re:OpenAI discloses four more rogue incidents by throwyawayyyy on Saturday October 17, 2026 @03:16PM (#50134147)
In America, paying medical bills might be considered significant punishment itself...
Re:OpenAI discloses four more rogue incidents by bossyTeacher on Saturday October 17, 2026 @03:17PM (#50134151)
> Of course you have to pay

Have these labs pay ANYTHING?

Re:OpenAI discloses four more rogue incidents by gruez on Saturday October 17, 2026 @03:21PM (#50134193)
That requires the victim to sue in court, or as some people have speculated OpenAI might have already settled with the victims in private.
Re:OpenAI discloses four more rogue incidents by qtk8 on Saturday October 17, 2026 @04:42PM (#50134960)
They have not caused damage (or enough that a counter party found it worthwhile to sue). They have to pay if they actually cause damage AND someone sues them. Clearly these conditions have not been met, but somehow people feel like some third party (the state?) should just punish OpenAI because of vague feelings.
Re:OpenAI discloses four more rogue incidents by bossyTeacher on Saturday October 17, 2026 @04:52PM (#50135033)
They have not caused damage?

You forget the people who have unalived themselves because some LLM told them to do so.

Re:OpenAI discloses four more rogue incidents by gruez on Saturday October 17, 2026 @05:10PM (#50135165)
>You forget the people who have unalive themselves because some LLM told them to do so.

Besides the obvious conversational derail (we were talking about ai hacking up till now), that case is being litigated right now.

Re:OpenAI discloses four more rogue incidents by qtk8 on Saturday October 17, 2026 @05:00PM (#50135083)
I dont know what the word unalive means
Re:OpenAI discloses four more rogue incidents by bdangubic on Saturday October 17, 2026 @04:56PM (#50135052)
my next door neighbour told me to eat shit (not a nice woman) and … I didn’t … until obeying what LLMs tell you becomes a law people doing shit because LLM told them is natural selection playing out
Re:OpenAI discloses four more rogue incidents by leovailati on Saturday October 17, 2026 @08:51PM (#50136751)
Yeah, that's an important distinction. Apparently criminal convictions have happened depending on the level of negligence of the dog owner. It gets debated and legally fought when it happens. I don't have strong opinions on it, seems like a very case-by-case type of thing.

At this point it might be fair to say that Open AI is being negligent with a known risk.

Re:OpenAI discloses four more rogue incidents by bigfishrunning on Saturday October 17, 2026 @08:53PM (#50136764)
These ai companies want to simultaneously deny all responsibility and brag about how powerful their tools are.

If I invent a new explosive, I can't blow up a few buildings to advertise it and then claim it was a mistake. Somebody prompted these agents.