[Altern4tiveAcc]

> Prosecutors say they are now investigating whether X has broken the law across multiple areas.

This step could come before a police raid.

This looks like plain political pressure. No lives were saved, and no crime was prevented by harassing local workers.

[bawolff]

> and no crime was prevented by harassing local workers.

Siezing records is usually a major step in an investigation. Its how you get evidence.

Sure it could just be harrasment, but this is also how normal police work looks. France has a reasonable judicial system so absent of other evidence i'm inclined to believe this was legit.

[moolcool]

> This looks like plain political pressure. No lives were saved, and no crime was prevented by harassing local workers.

The company made and released a tool with seemingly no guard-rails, which was used en masse to generate deepfakes and child pornography.

[trhway]

Internet routers, network cards, the computers, OS and various application software have no guardrails and is used for all the nefarious things. Why those companies aren't raided?

[sirnicolaz]

This is like comparing the danger of a machine gun to that of a block of lead.

[trhway]

May be. We do have codified in law definition of machine gun which clearly separates it from a block of lead. What codified in law definitions are used here to separate photoshop from Grok in the context of those deepfakes and CSAM?

Without such clear legal definitions going after Grok while not going after photoshop is just an act of political pressure.

[bootsmann]

Why do you think France doesn’t have such laws that delineate this legal definition?

What you’re implying here is that Musk should be immune from any prosecution simply because he is right wing, which…

[moolcool]

Take a step back and look at what you’re defending, man.

[trhway]

i don't understand why you're defending CSAM creation in photoshop.

[bluescrn]

They don’t provide a large platform for political speech.

This isn’t about AI or CSAM (Have we seen any other AI companies raided by governments for enabling creation of deepfakes, dangerous misinformation, illegal images, or for flagrant industrial-scale copyright infringement?)

[direwolf20]

No because most of those things aren't illegal and most of those companies have guard rails and because a prosecution requires a much higher standard of evidence than internet shitposting, and only X was stupid enough to make their illegal activity obvious.

[moorebob]

Deepfakes have been around long before X (and other chatbots) allowed undressing of real people.

The difference is that the entire political Left hate and fear Elon and are desperately trying to destroy him.

[trothamel]

Don't forget polaroid in that.

[pdpi]

I'm of two minds about this.

One the one hand, it seems "obvious" that Grok should somehow be legally required to have guardrails stopping it from producing kiddie porn.

On the other hand, it also seems "obvious" that laws forcing 3D printers to detect and block attempts to print firearms are patently bullshit.

The thing is, I'm not sure how I can reconcile those two seemingly-obvious statements in a principled manner.

[_trampeltier]

It is very different. It is YOUR 3d printer, no one else is involved. You might print a knife and kill somebody with it, you go to jail, not third party involved.

If you use a service like Grok, then you use somebody elses computer / things. X is the owner from computer that produced CP. So of course X is at least also a bit liable for producing CP.

[pdpi]

How does that mesh with all the safe harbour provisions we've depended on to make the modern internet, though?

[mikeyouse]

The safe harbor provisions largely protect X from the content that the users post (within reason). Suddenly Grok/X were actually producing the objectionable content. Users were making gross requests and then an LLM owned by X, using X servers and X code would generate the illegal material and then post it to the website. The entity responsible is no longer done user but instead the company itself.

[luke5441]

Yes, and that was a very stupid product decision. They could have put the image generation into the post editor, shifting responsibility to the users.

I'd guess Elon is responsible for that product decision.

[Altern4tiveAcc]

So, if someone hosts an image editor as web app, are they liable if someone uses that editor to create CP?

I honestly don't follow it. People creating nudes of others and using the Internet to distribute it can be sued for defamation, sure. I don't think the people hosting the service should be liable themselves, just like people hosting Tor nodes shouldn't be liable by what users of the Tor Network do.

[pjc50]

Note that is a US law, not a French one.

Also, safe harbor doesn't apply because this is published under the @grok handle! It's being published by X under one of their brand names, it's absurd to argue that they're unaware or not consenting to its publication.

[_trampeltier]

Before a USER did create content. So the user was / is liable. Now a LLM owned by a company does create content. So the company is liable.

[hbs18]

I'm not trying to make excuses for Grok, but how exactly isn't the user creating the content? Grok doesn't have create images on its own volition, the user is still required to give it some input, therefore "creating" the content.

[numpad0]

It's not like the world benefited from safe harbor laws that much. Why don't just amend them so that algorithms that run on server side and platforms that recommend things are not eligible.

[direwolf20]

If you are thinking about section 230 it only applies to user–generated content, so not server–side AI or timeline algorithms.

[jazzyjackson]

This might be an unpopular opinion but I always thought we might be better off without Web 2.0 where site owners aren’t held responsible for user content

If you’re hosting content, why shouldn’t you be responsible, because your business model is impossible if you’re held to account for what’s happening on your premises?

Without safe harbor, people might have to jump through the hoops of buying their own domain name, and hosting content themselves, would that be so bad?

[pdpi]

What about webmail, IM, or any other sort of web-hosted communication? Do you honestly think it would be better if Google were responsible for whatever content gets sent to a gmail address?

[direwolf20]

Any app allowing any communication between two users would be illegal.

[terminalshort]

You know this site would not be possible without those protections, right?

[ytpete]

The 3D printers don't generate the plans for the gun for you though. If someone sold a printer that would – happily with no guardrails – generate 3D models of CSAM from thin air and then print them, I bet they'd be investigated too. Or for that matter a 3D printer that came bundled with a built-in library of gun models you could print with very little skill...

[beAbU]

I don't have an answer, but the theme that's been bouncing around in my head has been about accessibility.

Grok makes it trivial to create fake CSAM or other explicit images. Before, if someone spent a week on photoshop to do the same, It won't be Adobe that gets the blame.

Same for 3D printers. Before, anyone could make a gun provided they have the right tools (which is very expensive), now it's being argued that 3D printers are making this more accessible. Although I would argue it's always been easy to make a gun, all you need is a piece of pipe. So I don't entirely buy the moral panic against 3D printers.

Where that threshold lies I don't know. But I think that's the crux if it. Technology is making previously difficult things easier, to the benefit of all humanity. It's just unfortunate that some less-nice things have also been included.

[_ph_]

I think a company which runs a printing business would have some obligations to make sure they are not fulfilling print orders for guns. Another interesting example are printers and copiers, which do refuse to copy cash. Which is partly facilitated with the EURion constellation (https://en.wikipedia.org/wiki/EURion_constellation) and other means.

[watwut]

Grok is publishing the CSAM photos for everyone to see. It is used as a tool for harassment and abuse, literally.

[pdpi]

Sure, and the fact that they haven't voluntarily put guard rails up to stop that is absolutely vile. But my personal definition of "absolutely vile" isn't a valid legal standard. So, the issue is, like I said, how do you come up with a principled approach to making them do it that doesn't have a whole bunch of unintended consequences?

[watwut]

Courts are able or should be able to distinguish "tool that creates an item in the privacy of your home" and "tool that disseminates nonconsensual pornographic picture to wide public". Legal standards with that level or definition are fairly normal.

[muyuu]

i don't see any need for guardrails, other than making the prompter responsible for the output of the bot, particularly when it's predictable

you cannot elaborately use a software to produce an effect that is patently illegal and accurate to your usage, and then pretend the software is to blame

[ljsprague]

No other "AI" companies released tools that could do the same?

[hackinthebochs]

In fact, Gemini could bikinify any image just like Grok. Google added guardrails after all the backlash Grok received.

[mekdoonggi]

And they should face consequences for that, somewhat mitigated by their good faith response.

[cubefox]

> The company made and released a tool with seemingly no guard-rails, which was used en masse to generate deepfakes and child pornography.

Do you have any evidence for that? As far as I can tell, this is false. The only thing I saw was Grok changing photos of adults into them wearing bikinis, which is far less bad.

[klez]

That's why this is an investigation looking for evidence and not a conviction.

This is how it works, at least in civil law countries. If the prosecutor has reasonable suspicious that a crime is taking place they send the so-called "judiciary police" to gather evidence. If they find none (or they're inconclusive etc...) the charges are dropped, otherwise they ask the court to go to trial.

On some occasions I take on judiciary police duties for animal welfare. Just last week I participated in a raid. We were not there to arrest anyone, just to gather evidence so the prosecutor could decide whether to press charges and go to trial.

[direwolf20]

Note that the raid itself is a punishment. It's normal for them to seize all electronic devices. How is X France supposed to do any business without any electronic devices? And even when charges are dropped, the devices are never returned.

[scott_w]

Did you miss the numerous news reports? Example: https://www.theguardian.com/technology/2026/jan/08/ai-chatbo...

For obvious reasons, decent people are not about to go out and try to general child sexual abuse material to prove a point to you, if that’s what you’re asking for.

[cubefox]

First of all, the Guardian is known to be heavily biased again Musk. They always try hard to make everything about him sound as negative as possible. Second, last time I tried, Grok even refused to create pictures of naked adults. I just tried again and this is still the case:

https://x.com/i/grok/share/1cd2a181583f473f811c0d58996232ab

The claim that they released a tool with "seemingly no guardrailes" is therefore clearly false. I think what instead has happened here is that some people found a hack to circumvent some of those guardrails via something like a jailbreak.

[scott_w]

For more evidence:

https://www.bbc.co.uk/news/articles/cvg1mzlryxeo

Also, X seem to disagree with you and admit that CSAM was being generated:

https://arstechnica.com/tech-policy/2026/01/x-blames-users-f...

Also the reason you can’t make it generate those images is because they implemented safeguards since that article was written:

https://www.ofcom.org.uk/online-safety/illegal-and-harmful-c...

This is because of government pressure (see Ofcom link).

I’d say you’re making yourself look foolish but you seem happy to defend nonces so I’ll not waste my time.

[cubefox]

> Also, X seem to disagree with you and admit that CSAM was being generated

That post doesn't contain such an admission, it instead talks about forbidden prompting.

> Also the reason you can’t make it generate those images is because they implemented safeguards since that article was written:

That article links to this article: https://x.com/Safety/status/2011573102485127562 - which contradicts your claim that there were no guardrails before. And as I said, I already tried it a while ago, and Grok also refused to create images of naked adults then.

[emsign]

> First of all, the Guardian is known to be heavily biased again Musk.

Says who? Musk?

[jibal]

That is only "known" to intellectually dishonest ideologues.

[Hikikomori]

>First of all, the Guardian is known to be heavily biased again Musk.

Biased against the man asking Epstein which day would be best for the "wildest" party.

[neoromantique]

>First of all, the Guardian is known to be heavily biased again Musk.

Which is good, that is the sane position to take these days.

[numpad0]

Grok do seem to have tons of useless guardrails. Reportedly you can't prompt it directly. But also reportedly they tend to go for almost nonsensically off-guardrail interpretation of prompts.

[317070]

Well, there is evidence that this company made and distributed CSAM and pornographic deepfakes to make a profit. There is no evidence lacking there for the investigators.

So the question becomes if it was done knowingly or recklessly, hence a police raid for evidence.

See also [0] for a legal discussion in the German context.

[0] https://arxiv.org/html/2601.03788v1

[skissane]

> Well, there is evidence that this company made and distributed CSAM

I think one big issue with this statement – "CSAM" lacks a precise legal definition; the precise legal term(s) vary from country to country, with differing definitions. While sexual imagery of real minors is highly illegal everywhere, there's a whole lot of other material – textual stories, drawings, animation, AI-generated images of nonexistent minors – which can be extremely criminal on one side of an international border, de facto legal on the other.

And I'm not actually sure what the legal definition is in France; the relevant article of the French Penal Code 227-23 [0] seems superficially similar to the legal definition of "child pornography" in the United States (post-Ashcroft vs Free Speech Coalition), and so some–but (maybe) not all–of the "CSAM" Grok is accused of generating wouldn't actually fall under it. (But of course, I don't know how French courts interpret it, so maybe what it means in practice is something broader than my reading of the text suggests.)

And I think this is part of the issue – xAI's executives are likely focused on compliance with US law on these topics, less concerned with complying with non-US law, in spite of the fact that CSAM laws in much of the rest of the world are much broader than in the US. That's less of an issue for Anthropic/Google/OpenAI, since their executives don't have the same "anything that's legal" attitude which xAI often has. And, as I said – while that's undoubtedly true in general, I'm unsure to what extent it is actually true for France in particular.

[0] https://www.legifrance.gouv.fr/codes/section_lc/LEGITEXT0000...

[direwolf20]

It wouldn't be called CSAM in France because it would be called a French word. Arguing definitions is arguing semantics. The point is, X did things that are illegal in France, no matter what you call them.

[skissane]

> It wouldn't be called CSAM in France because it would be called a French word. Arguing definitions is arguing semantics.

The most common French word is Pédopornographie. But my impression is the definition of that word under French law is possibly narrower than some definitions of the English acronym “CSAM”. Canadian law is much broader, and so what’s legally pédopornographie (English “child pornogaphy”) in Canada may be much closer to broad “CSAM” definitions

> The point is, X did things that are illegal in France, no matter what you call them.

Which French law are you alleging they violated? Article 227-23 du Code pénal, or something else? And how exactly are you claiming they violated it?

Note the French authorities at this time are not accusing them of violating the law. An investigation is simply a concern or suspicion of a legal violation, not a formal accusation; one possible outcome of an investigation is a formal accusation, another is the conclusion that they (at least technically) didn’t violate the law after all. I don’t think the French legal process has reached a conclusion either way yet.

One relevant case is the unpublished Court of Cassation decision 06-86.763 dated 12 septembre 2007 [0] which upheld a conviction of child pornography for importing and distributing the anime film “Twin Angels - le retour des bêtes célestes - Vol. 3". [0] However, the somewhat odd situation is that it appears that film is catalogued by the French national library, [1] although I don’t know if a catalogue entry definitively proves they possess the item. Also, art. 227-23 distinguishes between material depicting under 15s (illegal to even possess) and material depicting under 18s (only illegal to possess if one has intent to distribute); this prosecution appears to be have been brought under the latter category only-even though the individual was depicted as being under 15-suggesting this anime might not be illegal to possess in France if one has no intent to distribute it.

But this is the point - one needs to look at the details of exactly what the law says and how exactly the authorities apply it, rather than vague assertions of criminality which might not actually be true.

[0] https://www.legifrance.gouv.fr/juri/id/JURITEXT000007640077/

[1] https://catalogue.bnf.fr/ark:/12148/cb38377329p

[graemep]

> And I think this is part of the issue – xAI's executives are likely focused on compliance with US law on these topics, less concerned with complying with non-US law

True, but outright child porn is illegal everywhere (as you said) and the borderline legal stuff is something most of your audience is quite happy to have removed. I cannot imagine you are going to get a lot of complaints if you remove AI generated sexual images of minors, for example so it seems reasonable to play it safe.

> That's less of an issue for Anthropic/Google/OpenAI, since their executives don't have the same "anything that's legal" attitude which xAI often has.

This is also common, but it is irritating too as it means the rest of the world is stuck with silly American attitudes about things like nudity and alcohol - for example Youtube videos blurring out bits of Greek statues because they are scared of being demonetised. These are things people take kids to see in museums!

[krick]

To me, the most worrying part of the whole discussion is that your comment is pretty much the most "daring", if you can call it that, attempt to question if there even is a crime. Everyone else is worried about raids (which are normal whenever there is an ongoing investigation, unfortunate as it may be to the one being investigated). And no one dares to say, that, uh, perhaps making pictures on GPU should not be considered a crime in the same sense as human-trafficking or production of weapons are... Oh, wait. The latter is legal, right.

[orwin]

France prosecutors use police raids way more than other western countries. Banks, political parties, ex-presidents, corporate HQs, worksites... Here, while white-collar crimes are punished as much as in the US (i.e very little), we do at least investigate them.

[giancarlostoro]

> This looks like plain political pressure. No lives were saved, and no crime was prevented by harassing local workers.

I wouldn't even consider this a reason if it wasn't for the fact that OpenAI and Google, and hell literally every image model out there all have the same "this guy edited this underage girls face into a bikini" problem (this was the most public example I've heard so I'm going with that as my example). People still jailbreak chatgpt, and they've poured how much money into that?

[emsign]

They've already broken the law by creating and hosting CSAM. Now let's see what else prosecutors will find.

[bluegatty]

No, that's not at all how this works.

They have a court order obviously to collect evidence.

You have offered zero evidence to indicate there is 'political pressure' and that statement by prosecutors doesn't hint at that.

'No crime was prevented by harassing workers' is essentially non sequitor in this context.

It could be that that this is political nonsense, but there would have to be more details.

These issues are really hard but we have to confront them. X can alter electoral outcomes. That's where we are at.

[aaomidi]

Lmao they literally made a broad accessible CSAM maker.