Ireland 'not doing enough' to protect children from AI dangers 

Dáil committee hears children an 'afterthought' when it comes to developing policies around AI internationally.

Ireland is “not doing enough” to protect children from the dangers of artificial intelligence, despite having the “potential and the obligation to be world-leading in this space”, an Oireachtas committee has heard.

Caoilfhionn Gallagher KC, Ireland’s special rapporteur on child protection, said children had often been an “afterthought” when it comes to developing policies around AI internationally.

“Waiting for Europe means you’re behind the curve,” she told members of the Oireachtas children’s committee. “There’s an enforcement gap here.” 

Among those addressing the committee were members of the Irish Council for Civil Liberties and representatives from Cybersafe Ireland.

UCC's Professor Barry O’Sullivan noted how AI takes on many forms and represents both great opportunities and enormous risks.

“We’re in a very interesting year,” he said. “We’re now in the Super Bowl of democracy. In the next year, three billion people are going to the ballot box.

There’s going to be an insane amount of AI technology generating deepfakes on an hour, if not by-the-minute basis. We’re all going to be extremely concerned, and we’re going to be having a different conversation on this in six months’ time.” 

Cybersafe Ireland’s Clare Daly, meanwhile, said children were being “treated like guinea pigs in the digital world”.

“Nonconsensual pornographic deepfakes are alarmingly easy to access and create,” she said. 

“Deepfake software can take a person’s photos and face-swap them onto pornographic videos, making it appear as if the subject is partaking in sexual acts. 

"Research in this area points that whilst much of the abuse is image-based, it can also be used in grooming and sexual extortion text, which poses significant risks to children.” 

Green Party TD Patrick Costello asked the attendees if there was a liability on publishers such as the social media giants “for the harm that they’ve caused” and “if not, why not”.

Prof O’Sullivan said people who create content that is harmful should be accountable, while organisations that disseminate this content should also.

All in attendance agreed enforcement should be considerably stepped up.

Irish Council for Civil Liberties senior fellow Dr Johnny Ryan said there was a big deficit in this regard but Ireland had legislation available to take action, through mechanisms such as the GDPR.

“The powers are very broad,” he said. “We’ve got the Data Protection Commissioner and Coimisiún na Méan. We’ve the tools to solve these problems.

“If we had a a real culture of enforcement, dedicated investigation, [and] people actually willing to do the work of taking the scalp, we would be best practice because we have the law.” 

In terms of the development of AI, Ms Gallagher said children had been left out and child-centred design was required to protect young people online.

“The UN secretary general remarks on AI to the UN Security Council last July made zero mention on the rights of the child,” she said. 

“That’s quite a stark example of children being left out of the AI conversation.”