Legislation which could provide the template for regulating artificial intelligence in the UK is set to become EU law following approval by the European parliament today. MEPs voted by 523 to 46 to endorse an AI Act which will outlaw some uses of the technology while placing obligations on others. 

The act creates three tiers of AI regulation based on the perceived risk. Applications to be banned outright include biometric categorisation systems based on sensitive characteristics and untargeted scraping of facial images from the internet or CCTV footage. 'Emotion recognition in the workplace and schools, social scoring, predictive policing (when it is based solely on profiling a person or assessing their characteristics), and AI that manipulates human behaviour or exploits people’s vulnerabilities will also be forbidden,' the European Commission said. 

High-risk systems, including those in critical infrastructure, education, employment, essential private and public services including law enforcement will be subject to a new regulatory regime. Their operators will be required to assess and reduce risks, maintain use logs, be transparent and accurate, and ensure human oversight. Citizens will have a right to submit complaints about AI systems and receive explanations about decisions based on high-risk AI systems.

Artificial or manipulated images, audio or video content will have to be clearly labelled. The act also establishes an EU AI Office to supervise the implementation and enforcement of the act. 

Italian MEP Brando Benifei told the parliamentary debate: 'We finally have the world’s first binding law on artificial intelligence, to reduce risks, create opportunities, combat discrimination and bring transparency.'

The law is still subject to a final lawyer-linguist check and is expected to be finally adopted before the end of the legislature, the European Commission said. 

Its implementation will be watched closely in the US and UK. The British government last year published a white paper proposing a ‘pro innovation’ approach to regulating AI. 

However experts said that operators of AI systems should already take steps to align with the EU law. 'It is now crucial that businesses create and maintain a robust AI governance programme to make the best use of any AI technology and ensure compliance with the new regime,' said Marcus Evans, partner and European head of data privacy at global firm Norton Rose Fulbright. 'Businesses can expect more detail in the coming months on the specific requirements, as the EU Commission establishes and staffs the AI Office, and begins to set standards and provide guidance on the act.'

On a more sceptical note, Emma Wright, partner at media and technology firm Harbottle & Lewis, observed: 'The EU AI Act is the first significant attempt to regulate AI in the world – it remains to be seen whether the cost of compliance stifles innovation or whether the AI governance model that it establishes is a flagship export for the EU. Considering the pace of change in the technology – as shown with the launch of generative AI last year – a further complication could be that the EU AI Act quickly becomes outdated, especially considering the timeframes for implementation.'