After three years of work, Brussels presented on Wednesday its first draft of a legal framework to regulate artificial intelligence. “Artificial intelligence is already bringing us a lot, and this will increasingly be the case”, But “Excellence is only possible with confidence”, warned the Vice-President of the European Commission, Margrethe Vestager.
This will involve, for example, guaranteeing the safety of future autonomous vehicles or the respect of rights vis-à-vis surveillance systems. Generalized population surveillance systems, those used to manipulate the behavior, opinions or decisions of citizens, would be prohibited. Citizen rating systems, used in China in particular, would also be banned in Europe. Derogatory authorizations are however provided for the fight against terrorism, public security and military uses.
Applications classified according to their risks
The applications would be classified according to their risks. The most closely monitored would include remote biometric identification of people in public places and security features in critical public infrastructure. Compliance tests performed by a third-party authority are planned.
A second block would include systems for prioritizing emergency services, access to educational institutions, or recruitment tools. Their compliance will be guaranteed by a self-assessment conducted by the company, under the control of a supervisory authority. Finally, all the other uses, considered to be safe, would be subject to virtually no additional regulatory constraints.
Also to discover in video:
One billion euros per year
Europe has missed the consumer internet revolution, and has no champion equivalent to the five American giants Google, Apple, Facebook, Amazon and Microsoft, the famous Gafam, or their Chinese equivalents Baidu, Alibaba, Tencent and Xiaomi. . But nothing is lost concerning the revolution caused by the massive eruption of information technology in transport, energy, tourism, agriculture or medicine.
On the funding side, the EU plans to put on the table one billion euros per year for digital technologies, generating a total of 20 billion euros in investments over the next decade.
Towards generalized population surveillance?
The project was rather well received by the companies, for others, the text is insufficient. Proposal “Opens the door to general surveillance under the guise of public security”, denounces environmental MEP David Cormand. “Too many problematic uses of technology are allowed, such as the use of algorithms to predict crime”, says Orsolya Reich of the NGO Liberties.
The text will be debated for several months with the European Parliament and the 27 Member States, before the implementation of a final version.