European Union moves closer to passing the world's first law governing AI

The EU took a significant step towards enacting one of the world's first laws controlling artificial intelligence when its main legislative arm approved the language of draught legislation that includes a blanket prohibition on police use of live face recognition equipment in public places.

The European Parliament passed guidelines to establish a worldwide standard for the technology, including anything from automated medical diagnoses to certain types of drones, AI-generated movies known as deep fakes, and chatbots like ChatGPT.

MEPs will now negotiate specifics with EU member states before the drafting guidelines, known as the AI Act, become law.

"AI raises numerous social, ethical, and economic concerns." But there is a better time to press any 'pause' buttons. On the contrary, it is about moving quickly and taking responsibility," said Thierry Breton, the European Commission's internal market commissioner.

A revolt by Centre-right MEPs in the EPP parliamentary grouping against an absolute ban on real-time facial recognition on European streets did not materialize, with several parliamentarians attending Silvio Berlusconi's funeral in Italy.

The final result counted 499 in support, 28 against, and 93 abstentions.

European Union officials are anticipated to oppose a comprehensive ban on biometrics, with police forces eager to capitalize on the ability to identify offenders as they walk down the street or through public spaces.

The EPP said the technology might be critical in combatting crime, counter-terrorism intelligence, and searching for missing children.

Emotional recognition—used in some parts of China to identify tired truck drivers—will also be outlawed under the new regulation at work and in schools.

Roberta Metsola, president of the European Parliament, praised it as "legislation that will undoubtedly set the global standard for years to come." She stated that the EU now has the potential to set the tone globally and that "a new age of scrutiny" has begun.

Brando Benifei, a co-rapporteur of the parliament's AI committee, which advanced the legislation to the voting stage, stated that the measure would give "a clear safeguard to avoid any risk of mass surveillance" for face recognition.

Dragos Tudorache, another co-rapporteur, stated that if the laws had previously been in place, the French government would have needed to adopt legislation this year enabling live face recognition for crowd monitoring during the 2024 Olympics.

To fight the significant danger of copyright infringement, the legislation would require AI chatbot developers to share all works used to train them by scientists, singers, artists, photographers, and journalists. They must also demonstrate that everything they did to teach the machine was legal.

If they do not, companies may be required to remove applications immediately or face a punishment of up to 7% of their sales, which could amount to hundreds of millions of euros for tech behemoths. "There are plenty of sharp teeth in there," Tudorache explained.

He stated that negotiations with the European Council and the European Commission would begin and that he would approach them with a mandate from parliament rather than "red lines" on the contentious topic of face recognition.

Benifei characterized the EPP's efforts to overturn the blanket prohibition on mass surveillance because it would prevent police from utilizing the instrument for security as propaganda. After all, authorities would still be allowed to use biometric data, including CCTV video, to pursue criminals, as they do now.

There is also a rising need for AI regulation on both sides of the Atlantic as pressure mounts on Western governments to act quickly in what some view as a war to save humankind.

While supporters of AI applaud the technology's promise to alter society, including labor, healthcare, and artistic endeavors, others are concerned about its potential to harm democracy.

Even if the EU meets its ambitious goal of reaching an agreement on the law by the end of the year, it will not enter into force until at least 2026, prompting the EU to seek a voluntary interim arrangement with internet corporations.