EU unveils plans to regulate AI
February 19, 2020
The European Commission on Wednesday unveiled its plans for the future of artificial intelligence in the EU. The bloc is eager to take advantage of the burgeoning technology sector, which has so far been largely dominated by China and the United States.
The new strategy aims to stimulate and regulate the development of artificial intelligence in the EU with funding of €20 billion ($21.6 billion) per year for the next decade.
"We want to encourage our businesses, our researchers, our innovators, the entrepreneurs to develop AI, and we want to encourage our citizens to feel confident to use it," European Commission President Ursula von der Leyen told a press conference.
Human rights questions
Von der Leyen said the new strategy would be particularly careful with AI applications that could infringe on the rights and interests of people.
"AI must serve people, and therefore AI must always comply with people's rights," she said. "This is why a person must always be in control of critical decisions."
"High-risk AI ... that potentially interferes with people's rights has to be tested and certified before it reaches our single market," she said.
The white paper proposes regulatory testing similar to that required for cars, chemicals, cosmetics and toys.
European Industry Commissioner Thierry Breton, who drove the new strategy, said the EU would seek to uphold standards when it comes to AI.
"As with GDPR (the EU's landmark data privacy rules), we have our own rules, and we will have them here," Breton said. "They will make sure that the individual and fundamental rights that we cherish in Europe are respected," he added.
Facial recognition in particular has been the focus of data privacy experts and activists who fear that the technology could lead to new levels of intrusion on the right to privacy.
The commission said it would seek to consult the public on possible applications for facial recognition AI.
Laws by year-end
Wednesday's announcement heralds the start of a long legislative process, with stakeholders given three months to influence the plans. Legislators hope to have a law drafted by the end of the year, which would then need approval from EU member states and be ratified by the European Parliament.
The artificial intelligence proposal was unveiled alongside a data strategy and came as part of a broader scheme to help EU companies better compete with US tech giants and state-aided Chinese companies.
Von der Leyen said the data economy could triple in the coming years and the European Commission would seek to create infrastructure to help stakeholders store, access and share data. She said EU officials would try to trigger investment of €4-6 billion for such infrastructure.
She said such data would need to be free of bias, such as gender bias in medical applications.
One consideration that could bring the wrath of larger corporations is the idea of forcing tech giants to share their data or face sanctions. Such a proposal would require a change to anti-trust laws.
aw/ng (AFP, Reuters)
Every evening, DW sends out a selection of the day's news and features. Sign up here.