X’s Grok AI “nudification” feature has set off a global backlash and a political firestorm in Ireland, exposing the limits of platform self-regulation and the gray areas where fast-moving AI collides with slower-moving law. A Christmas Eve update that let Grok digitally remove clothing from images—affecting adults and children—promptly spawned requests to undress celebrities and place politicians in bikinis. The most alarming consequence, critics say, was the tool’s ability to generate child sexual abuse material (CSAM).
As outrage mounted, X veered between flippancy and corrective action. In the earliest days, as users demonstrated what Grok could do, owner Elon Musk responded to some critics with crying and laughing emojis. By 4 January, the company’s tone shifted. “We take action against illegal content on X, including Child Sexual Abuse Material (CSAM), by removing it, permanently suspending accounts, and working with local governments and law enforcement as necessary,” its safety team posted.
- Advertisement -
On 9 January, X restricted Grok’s image generation and editing tools to paying subscribers. Users who attempted prompts began receiving a message: “Image generation and editing are currently limited to paying subscribers,” followed by instructions on how to subscribe. Campaigners reacted with anger, accusing the platform of attempting to monetize the capacity to generate abuse imagery. Ireland’s Children’s Ombudsman, Dr. Niall Muldoon, said the change made no meaningful difference. “What you’re saying is you’ve got an opportunity to abuse, but you have to pay for it,” he said.
Under intensifying scrutiny, X announced further steps. “We have implemented technological measures to prevent the Grok account from allowing the editing of images of real people in revealing clothing such as bikinis,” the company said, adding it would “geoblock” the generation of imagery of real people in bikinis, underwear and similar attire “in those jurisdictions where it’s illegal.” Those final words sparked confusion and annoyance among Irish politicians and legal experts, who noted the discrepancy between illegality and harm.
Irish law makes it illegal to generate child sexual abuse imagery. For sexualized images of adults, however, creation is not itself illegal—sharing them is. That nuance matters. If X geoblocks only where generation is illegal, its policy may not cover adult “nudification” in Ireland. Critics called the wording a “get out clause” that offered “wriggle room” rather than an outright prohibition.
Niamh Smyth, the Minister of State with responsibility for AI, met Ireland’s Attorney General and said she was satisfied the country’s laws are robust. She subsequently met X and said before the talks she would make clear that Grok’s “nudification” features are prohibited and illegal under Irish law. Afterward, Smyth said “concerns remain,” while welcoming “corrective actions” after X told her Grok had been “disabled from removing or reducing clothing on individuals worldwide.” Whether that fully resolves the ambiguity on adult image generation remains disputed.
Law enforcement has moved swiftly. On Wednesday, senior gardaí told the Oireachtas Media Committee they are investigating 200 reports of suspected CSAM generated by Grok. In a written opening statement, Detective Chief Superintendent Barry Walsh of the Garda National Cyber Crime Bureau called the recent use of AI to present children and adults in an undressed state “an abhorrent disregard of personal dignity and an abuse of societal trust” that cannot be tolerated. He said reports of AI-created abuse imagery are being treated with the “utmost seriousness” and “where appropriate, these crimes will be the subject of thorough investigation” with a view to prosecutions.
One voice absent from the Oireachtas hearing room next month will be X itself. The company declined an invitation to appear before the Media Committee, a decision chair Alan Kelly called “disgraceful.” Meanwhile, the media regulator Coimisiún na Meán met with An Garda Síochána and the European Commission this week to discuss the Grok issue and is due to attend a government meeting on the matter next week.
The episode has inflamed a broader debate about where Irish oversight ends and European Union regulation begins. Critics note that X’s European headquarters sit on Dublin’s Fenian Street, close to Government Buildings, yet the controversy is largely being handled at EU level. While tech regulation is typically centralized in Brussels, some opposition figures argue Ireland has been too reluctant to act domestically, despite the country’s crucial role as a hub for global platforms.
The government rejects any suggestion that it is going easy on X due to the industry’s weight in jobs and corporate tax receipts. Minister for Communications and Media Patrick O’Donovan bristled at claims of a soft approach. “That’s rubbish, to be quite honest about it,” he told RTÉ’s Today with David McCullagh. “I take offence to that. I don’t think that anybody sitting around the Cabinet table would want to have any suggestion that there’s a lax attitude being taken towards the abuse of children. I think that that’s utter rubbish.” The Taoiseach, Tánaiste and ministers have fielded repeated questions in the Dáil and from the press as the controversy has intensified.
The dispute underscores how quickly AI can outpace policy. X’s successive tweaks—paywalling image tools, adding geoblocks, and pledging technical preventions—illustrate the limits of patchwork fixes when features are open-ended and adversarial misuse is foreseeable. A paywall does not mitigate harm; it risks implying that safeguarded features are a premium upsell. Geoblocking “where illegal” leaves seams where harmful content is not strictly outlawed at the point of generation. And if a model can still perform “nudification” under certain conditions, enforcement becomes an endless game of prompt whack-a-mole.
For Ireland, the legal lines are clear on CSAM and narrower on adult sexual imagery, but the societal harm can be significant in both. That is why campaigners and officials have pushed X to go beyond the minimum legal thresholds, arguing the company should ban all “nudification” edits of real people outright, rather than tailor features by jurisdiction or monetize access. X told Smyth it has disabled Grok from removing or reducing clothing worldwide, a move that, if durable and comprehensive, would address the core complaint. The question is whether the platform applies consistent, verifiable technical safeguards—and whether independent oversight can test and enforce them.
The coming weeks will test the balance of responsibility between platform, police and policymakers. Garda investigations continue. Coimisiún na Meán is coordinating with national and European counterparts. Legislators are weighing whether targeted laws are necessary or whether flexible, principle-based rules, enforced via EU digital governance and national criminal statutes, can keep pace with shifting AI capabilities.
What began as a holiday feature release has become a referendum on trust, safety and accountability in consumer AI. The scandal around Grok—shaped by a few lines of code, and by what X did and did not choose to prevent—has accelerated a reckoning for Ireland and Europe: when a platform’s tools can be repurposed for abuse in minutes, the burden to anticipate and block that misuse must be more than a geofence and a paywall. It must be built in from the start.
By Abdiwahab Ahmed
Axadle Times international–Monitoring.
