EU parliament approves groundbreaking regulations on AI

Avatar photo

By Rhys Duncan on

Potential framework for UK to follow

The European Parliament has endorsed a new legal framework concerning the use of artificial intelligence (AI), creating a potential pathway for the UK to follow.

After MEPs voted overwhelmingly in favour of the new laws earlier this week, 523 to 46, certain AI uses look set to be banned, with permitted tech to come under varying levels of scrutiny.

“All AI systems considered a clear threat to the safety, livelihoods and rights of people will be banned, from social scoring by governments to toys using voice assistance that encourages dangerous behaviour,” an article on the European Commission’s site reads.

For those programmes and uses that fall below this level, the legislation plans to divide AI into three tiers; high risk, limited risk, and minimal risk.

The high risk category includes AI used in critical infrastructure, education, safety components, employment processes, law enforcement, migration, and the administration of justice. For AI to be permitted in these areas it must first meet a number of obligations focussed on risk assessment and mitigation, security, and human oversight.

The 2024 Legal Cheek Firms Most List

Limited risk focuses on the dangers associated with a lack of transparency in AI usage. These uses will again be subject to obligations, albeit less stringent and focussed on ensuring that people are aware of the use of AI, for example with the labelling of AI chatbots and AI generated text, video, and images.

Those uses considered minimal risk will be unaffected. This category, the Commission says, covers “the vast majority of AI systems currently used in the EU”, and includes AI video games and spam filters.

This comprehensive framework would be the first of its kind, and could provide a pathway for the UK to follow.

Earlier this week the Solicitors Regulation Authority issued a warning to lawyers over the potential fabrication of deepfake clients using AI. This followed previous guidance highlighting the benefits and risks of AI given to solicitors, barristers, and judges over recent months.

Related Stories

Beware of ‘deepfake’ clients, regulator warns lawyers

Concerns over money laundering and terrorist financing

Mar 13 2024 7:53am

Shoosmiths advises TC seekers on using AI in applications

Refine original thoughts, not replace them

Feb 22 2024 8:14am