[slopinthebag]

Not just the users, the service providers too! If I go to any other business and pay them to break the law and they do it, they're also liable! If you ask OpenAI or xAi to break the law and they do it, why shouldn't they also be responsible?

[fragmede]

Do we hold gun manufacturers responsible for the deaths from their guns? The answer to that Isa whole quagmire that is basically the same.

[dredmorbius]

If you're referring to the United States, the answer is that yes, weapons manufacturers were being held liable for firearms-based violence, until passage of the "Protection of Lawful Commerce in Arms Act" of 2005:

<https://en.wikipedia.org/wiki/Protection_of_Lawful_Commerce_...>

(Note that the bill passed with significant Republican and Democratic support.)

Which is to say, conventional notions of contributory negligence and liability existed until sufficient political power was accrued by the arms industry to write itself out of same. The cited Wikipedia article names several such suits, notably by the cities of Chicago, IL, and Bridgeport, CT.

[ponector]

You should compare not with manufacturers but with gun resellers. And yes, they may bear the responsibility.

Tighter regulation for AI is needed.

[danaris]

I think there's a meaningful difference between a gun manufacturer and an LLM service provider like OpenAI.

A better analogy would be some kind of "guns-as-a-service" model, where the company sends down a drone with a gun and fires it at whoever you point it at, then the drone flies back to base.

I think it would be very clear in cases like that that the service provider should be held liable.

[tremon]

The parent said service provider, not manufacturer. You are effectively claiming that hitmen should not be prosecuted, only the person that hired them.

[JumpCrisscross]

> Do we hold gun manufacturers responsible for the deaths from their guns?

In a lot of the world, yes, and in America we would as well if it weren’t for the modern take on the Second Amendment. AI has no similar legal purchase.

[card_zero]

But there are lots of similar quandaries:

Bitey dogs.

Dangerous drugs and their users and purveyors. Heroin, weed, booze, coffee.

Things done while on drugs. Things done while insane.

Unhealthy food and its purveyors and consumers.

Social media and its "addicts". TV, any old media, and social panic.

The question "whose fault?" isn't simple.

[JumpCrisscross]

> * The question "whose fault?" isn't simple*

Most of your examples have exquisitely-simple causation.

[card_zero]

OK, I'll say "no they don't". Over to you.

[JumpCrisscross]

Literally first one. “Bitey dogs.” In what jurisdiction is a dog biting a person legally ambiguous in respect of liability?

[lwelyk]

AI training and algorithms are trained and guided to certain kinds of results. Grok, for example, is claimed to be modified constantly according to Elon Musk's whims.

If you theoretically trained an AI on libel and had it set to libel anyone at the slightest prompt, then allowed users to make a request that had your AI on your server use your services to libel someone, I'm not really seeing how you would not be liable.

[slopinthebag]

I never said the model developers should be responsible. I said service providers. If someone downloads a local model and breaks the law, the responsibility is solely on the user. But if someone uses a service provider to break the law, that service provider is obviously partially responsible, since they literally fulfilled the illegal request.