EU Lawmakers Vote to Completely Ban Remote Facial-Recognition Surveillance

Real-time, remote biometric monitoring has been outlawed by the European Parliament, making it impossible for police departments around the EU to employ live facial recognition technology. Legislators had earlier approved the restriction after confronting early resistance over worries that it would be overly broad.

The restriction might put the parliamentarians in conflict with EU countries which would rather heavily implement the technology for policing.

The decision also includes further restrictions on general-purpose artificial intelligence (AI) and so-called fundamental models, such OpenAI’s GPT-4. Regardless of their intended use, the law mandates that businesses like OpenAI Inc. and Google do risk assessments and give descriptions of the copyrighted content used to train their models.

Roberta Metsola, the president of the parliament, stated during a news conference that any upcoming advancements in artificial intelligence would need to be subject to consistent, clearly stated rules and constraints.

“There is one thing that we will not compromise on: Anytime technology advances, it must go hand in hand with our fundamental rights and democratic values,” he added.

On June 14, the AI Act’s whole draft was approved by a resounding majority. 499 delegates voted in favor, 28 voted against, and 93 did not participate in the voting. The vote kicks off what is known as the “trilogue” round of discussions amongst the parliament members, where the finer points of the measure will be finalized.

According to Bloomberg, the EU Commission is aiming for a deal by the end of the year, meaning the new AI Act regulations could affect businesses as soon as 2026.

However, according to Brando Benifei, one of the primary writers of the measure, some regulations may be put into place much earlier.

In the meanwhile, representatives from the EU, notably Executive Vice President Margrethe Vestager, are working to develop a voluntary code of conduct with G7 digital corporations that also includes countries like India and Indonesia.

Facial scanning in public places would be authorized for specific law-enforcement scenarios under an earlier agreement reached by EU member states at the end of last year. This issue will continue to be a sticking point for a number of EU member states in the next discussions.

People’s Party members who lean farther to the right petitioned to add exceptions to the rule, such as in cases of looking for missing children and stopping terrorist acts. But in the resounding vote earlier this month, legislators mostly ignored these arguments.

“The result of today gives us even a stronger position. It’s clear that the parliament doesn’t want us to recede on such important topics, on avoiding mass surveillance,” said Benifei.

Concerns

In its report (pdf), the European Parliamentary Research Service describes how AI technology such as facial recognition could be problematic in terms of personal freedom.

“While there are real benefits to using facial recognition systems for public safety and security, their pervasiveness and intrusiveness, as well as their susceptibility to error, give rise to a number of fundamental rights concerns with regard, for instance, to discrimination against certain segments of the population and violations of the right to data protection and privacy,” the report states.

The AI Act was first proposed in 2021 (pdf), and was initially touted as a risk-based approach that would effectively regulate the application of AI rather than the actual technology itself. The proposal would have banned practices like social credit scoring and the implementation of technologies that would likely have negative impacts in terms of bias, discrimination, and citizens’ fundamental rights.

The drive to include general-purpose AI in the act, which would apply to a considerably larger range of circumstances, was instead led by EU member states.

According to reports, members of the European Parliament went a step further by putting restrictions to “foundational models.” This refers to the substantial language models that underlie chatbots like ChatGPT, which have recently been under intense public and official attention.

In a statement, Internal Market Commissioner Thierry Breton said he expects clear and proportionate rules on generative AI to be focal points of the trilogue process.

“We need effective transparency requirements on AI-generated content and strict rules against ‘deep fakes,’” he said.

The scope of the new EU legislation may have significant effects on a market that is now pegged at close to $1.5 trillion in value. If they don’t comply, businesses might be subject to hefty fines of up to 6% of their annual income.