What The UK AI Lawsuit Against Musk Reveals

Via ABC News:

A British lawmaker said Thursday she is suing Elon Musk’s company xAI for invasion of privacy, alleging that fake images of her were created using the Grok chatbot.

Jess Asato, a legislator with the governing Labour Party, says someone used Grok to create fake images of her in a bikini without her consent in January after she criticized the spread of deepfake pornography online.

She filed a claim Wednesday at the High Court in London, citing misuse of private information under the Data Protection Act.

Just looking at her headshot above, I'm not sure I would want to see a photo of her in a bikini, fake or real. But just as a US observer, other things strike me: first, you can create a picture of anyone in a bikini just with Photoshop, no need to use AI, and people do it all the time. It's ridiculous to think about suing Photoshop because someone can create a phony picture of someone in a bikini. Second, at least in the US, Ms Asato is a public figure, and people have broad protections for satirical impressions of public figures.

But that's where the real problem starts, this is the UK, not the US. I asked the AI oracle, "UK MP Jess Asato is suing xAI because someone had it make a picture of her in a bikini. How would this suit fare in the US?" It answered,

If UK MP Jess Asato filed her lawsuit against xAI in the United States, the suit would almost certainly face dismissal due to powerful tech liability immunities, strict free speech protections for public figures, and a fragmented patchwork of state deepfake laws. While her UK suit relies on specific Data Protection Act provisions and the tort of "misuse of private information," a US federal or state court would analyze the case through radically different legal frameworks.

The biggest hurdle for her case in the US would be Section 230 of the Communications Decency Act. Section 230 protects interactive computer services from being treated as the publisher or speaker of content provided by a third-party user.

As an elected Member of Parliament, Asato is a high-profile public figure. If she pursued standard US torts like Defamation or Intentional Infliction of Emotional Distress (IIED), she would hit a constitutional wall:The "Actual Malice" Standard: Under US law, public figures must prove a platform acted with "actual malice"—meaning xAI knew the images were false and defamatory and distributed them with reckless disregard for the truth. Because a chatbot automatically generates images based on random user inputs, proving the corporate entity itself harbored subjective malice toward her is nearly impossible.

US law heavily protects parodies or caricatures of political figures, making standard privacy or likeness claims much harder to win for politicians than for private citizens.

In other words, her suit would be laughed out of court in the US. Worse, it would trigger the Streisand Effect, wherein public curiosity would be whetted, and people would want to see the picture that caused all the controversy, with the result that it, and viral takes on it, would be all over the web. But remember, this is the UK, not the US. I asked the AI oracle, "Is the picture of UK MP Jess Asato in a bikini available on the web?" It answered, wagging its virtual finger:

No real picture of UK Member of Parliament (MP) Jess Asato in a bikini exists, as the images circulating online are fake, AI-generated deepfakes created without her consent.

Major social media platforms and mainstream media outlets have banned and actively removed these explicit images from the web. However, because users continue to try to re-upload them maliciously, some copies or variations might still exist in hidden or unmoderated corners of the internet.

So naturally, I went searching the web to see if the fake photo is extant, at least outside the UK, and I don't know if what I found is the authentic fake or a fake fake, but it's, er, what you'd imagine, or as someone characterized it, "a hippo in a string bikini". But now we've got a problem: what if a fake version of the fake photo goes viral and reaches the UK? Can those who upload the fake fake version claim they can do it with impunity because it's not the real fake version?

I asked the AI oracle,

Why does the UK have so little legal protection for freedom of speech?

It answered,

The United Kingdom does protect free speech, but it treats it as a "qualified right" rather than an absolute rule. This means the government balances your right to speak with the public's right to safety, privacy, and protection from harm.

Because of this, the UK has stricter laws than countries like the United States. If you say something that causes severe harm or breaks the peace, the government can step in and punish you.

The UK does not have a single, written constitution.

The United States has the First Amendment. It clearly says Congress cannot make laws to stop free speech.

The UK relies on a loose system of historical documents, court decisions, and laws passed by Parliament. Because Parliament has ultimate power, it can change or add limits to speech whenever it feels it is necessary.

In other words, there's no UK equivalent to the wording in the US Constitution that "Congress shall make no law. . .", and Parliament can make any law it pleases. This is at the root of the current Labour program:

Protecting What Matters, a document outlining Labour’s new cohesion strategy, was officially unveiled by the government this evening after its contents were leaked to The Spectator on Friday.

. . . Steve Reed, the Communities Secretary, said: “We will not allow hatred to destroy the lives and life chances of those who are targeted.

. . . He said: “Cohesion underpins our economic strength, our democratic freedom and our national security. It is a fundamental part of the Britain we love. We have made our choice in place of division, we choose unity, and we know the people of Britain have made the same choice.”

Last week it was reported a special representative on anti-Muslim hostility would also be central to the government’s plans. The plan will include measures to tackle religious hatred and racial discrimination and clamp down on extremism in charities and universities.

. . . Lord Young of Acton, General Secretary of the Free Speech Union, told PoliticsHome he was concerned the plan will be used to “enforce radical progressive dogma”.

He said: “First, I worry that free speech will be curtailed in the name of promoting social cohesion – that it’s a synonym for social control. Second, I worry that measures designed to counter extremism will be applied to the Right but not to the Left, which is what’s happened with Prevent.

Of course it will, that's the point. Labour, having won a majority in the 2024 election, doesn't face another until 2029, notwithstanding it's become highly unpopular, since there's no UK equivalent to US midterm House elections, and it will simply continue to impose its program until it can't.

A major threat to Labour's control of speech in the UK is Elon Musk. He says what he pleases from the US, but his reach is instant and worldwide via his platform X, and there's little Labour can do about it -- except to sue one of his companies because someone made a fake photo of a frumpy middle-aged Labour MP in a bikini. Big threat. Take that, Elon Musk!