Minister of State with responsibility for ArtificiaI Intelligence Niamh Smyth is due to meet representatives of the social media platform X to discuss its measures to tackle the generation of sexualised images of women and children by its AI chatbot Grok.

Minister Smyth held a meeting with the Attorney General yesterday to ensure there were no legal loopholes for X or any other platform to allow AI tools to undress images of real people.

Speaking on RTÉ’s Six One, Ms Smyth said the purpose of her meeting with the AG was to “give myself and the Government absolute confidence in the fact that the laws that we have in place here in Ireland do protect our citizens in every way possible”.

Yesterday, X confirmed it will “geoblock” the ability of all Grok and X users to create images of people in “bikinis, underwear, and similar attire” in jurisdictions where such actions are deemed illegal.

The move follows widespread outrage over features that allowed users to digitally undress images of people.

Labour TD Alan Kelly confirmed that X has been invited to appear before the Oireachtas Committee on Media on 4 February.

It comes as a senior counsel specialising in media, data protection and commercial law has called on the European Commission to work with Coimisiún na Meán to take EU-wide action against AI apps that generate sexually explicit images.

The prompt screen from the Grok AI app is displayed on an iPad
X has been invited to appear before an Oireachtas committee regarding its AI chatbot Grok

Speaking on RTÉ’s Morning Ireland, Ronan Lupton said this approach would mean that action taken would be comprehensive, rather than being “a patchwork” of different measures across the EU.

Mr Lupton said the best possible outcome is that Coimisiún na Meán, specifically the Internet Safety Commissioner and the Digital Services Content Commissioner, work closely with the European Commission.

Mr Lupton said that interim measures are required to ban Grok from allowing nudification and allowing content to be created by users “who are effectively acting nefariously”.

He added that generating explicit sexual images of people without their consent is against the law.

However, authorities would have to discover evidence in order to prosecute offenders, he said, adding that there are “civil remedies” that could also be used to tackle the issue.

Mr Lupton said there are other AI apps that can generate sexually explicit images that could be breaking the law.

“We have a category of position which is lawful but awful, and that may be the content is subject to perhaps standards which may not breach the law fully, but ultimately if it does things which fundamentally gives rise to harassment or other forms of criminal offence, then it’s breaking the law.”

He added that his view is that the social media platform X has been “in decay” since being purchased by Elon Musk and that the images generated by Grok AI represent “deleterious content and bad for society”.

“I’m not saying that as someone who believes that we should take a sledgehammer to X or any other platform for that matter.

“Freedom of expression is a very important right,” he said.