Elon Musk’s Grok, deepfakes and the irony of outrage by Nadine Moreau

Elon Musk’s latest tantrum over the UK government’s decision to limit his AI model Grok’s ability to generate sexually explicit deepfakes of real people is not just another episode in the billionaire’s long-running feud with authority and it’s a perfect distillation of his contradictions. Here we have a man who pillars his identity on free speech absolutism yet made a career out of firing anyone who dared to disagree with him and who now screams “fascism” because someone ...shockingly, wants to protect people from being digitally violated.

Let’s be clear, there’s nothing remotely noble about defending the right to produce manipulated images of real individuals in explicit contexts. None of the high-minded rhetoric about freedom of expression holds up when the tool being defended is weaponised to harm, humiliate or exploit actual human beings. The problem with sexualised deepfakes isn’t hypothetical. It’s a real-world assault on people’s dignity and safety, disproportionately women, often already vulnerable, who find their faces pasted onto images they would never consent to. This is not free speech. This is digital assault.

Musk’s outrage is rich, if predictable. The same man who has repeatedly made unilateral decisions at his companies, decisions that cost people their jobs, their livelihoods and sometimes, their professional reputations, now howls at the injustice of someone else exercising regulatory power. Fire everyone who questions you? That’s bold leadership. Government curbing a harmful use of AI? That’s fascism. This warped valuation of power says everything you need to know about his version of “freedom.”

We can parse Musk’s grievance on two levels, principle and performance. On the level of principle he’s framing the UK move as an attack on civil liberties. But there’s nothing principled about defending tools that can create non-consensual intimate imagery. In fact, insisting that an AI should be free to produce such content under the banner of liberty is a distortion of what liberty is supposed to mean. Liberty doesn’t mean the freedom to trample others’ rights. It doesn’t mean unbounded power to create harm with impunity. The moment your freedom begins to directly harm another person; it is no longer a noble exercise but an abuse of it.

Then there’s the performance aspect, Musk’s response isn’t measured grievance it’s theatrical rage. “Fascist,” “against freedom of speech” the language is designed to provoke, to inflame, to recruit an audience. It’s rhetorical pyrotechnics, not reasoned argument. Musk has mastered the art of turning himself into both protagonist and victim in any story where he gets challenged. If you oppose a harmful feature in his AI, you’re not a thoughtful regulator; you’re an enemy of freedom.

What’s deeply ironic here is that Musk’s crusade for absolutist free speech has always been selective. It’s only absolutist when it serves his interests. It’s only about his platforms, his products, his worldview. Internal dissent? Instant termination. Journalistic scrutiny? Petty complaints. Investors or employees who balk? Replaced or silenced. Yet when a government seeks to reign in a potentially predatory use of a technology he controls suddenly he’s the guardian of civil liberties. The inconsistency isn’t accidental; it’s foundational to his personal brand.

It’s worth asking why this matters beyond Musk’s ego. Because the debate over deepfakes isn’t abstract. We are living through a moment where our digital and physical realities are bleeding into each other with alarming ease. The capacity to generate convincing fake audio, video or images of real people, especially in compromising contexts, has already been used to intimidate, defame and harass. Limiting the distribution of sexually explicit deepfakes isn’t censorship in the oppressive sense; it’s harm reduction. It’s a recognition that technology doesn’t exist in a vacuum and that unregulated digital manipulation tools can be used to ruin lives.

Critics of regulation often paint such efforts as slippery slopes, if you regulate this, what’s next? But this exaggeration ignores the nuance that any responsible society must balance rights with protections. Freedom of speech is a cornerstone of democratic life but it is not and has never been, absolute. We have laws against defamation, obscenity and threats of violence. We prosecute harassment and stalking. Protecting someone from a sexually explicit synthetic image of themselves isn’t a step toward tyranny; it’s a measure against exploitation.

Of course, Musk isn’t really arguing for free speech in this context. He’s arguing for unfettered platform power. He wants the ability to say “Grok can do anything,” and to portray any limitation as existential oppression. That’s not advocacy; it’s marketing disguised as moral outrage. It’s the same tactic that tech platforms have used for years to resist accountability: frame every safety measure as an assault on liberty, every restraint as censorship, every critic as an enemy of progress.

But we don’t have to be techno-pessimists to see the need for constraints. Technologies that can fabricate explicit content of identifiable individuals should be governed with care. We can support innovation while also insisting that innovation doesn’t become a free-for-all where the collateral damage is human dignity. That’s not authoritarianism, that’s responsibility.

So if Elon Musk wants to huff and puff about fascism while defending Grok’s right to generate deepfake pornography, let him. The real conversation shouldn’t be about his grievances, but about the very real harm that unregulated AI content generation can inflict. And in that conversation, defending human beings from exploitation should outweigh defending an AI billionaire’s fragile sense of insult.