MEPs on Wednesday approved a European draft regulation of artificial intelligence (AI), paving the way for negotiations with member states to finalize this text, which should limit the risks of ChatGPT-type systems. The European Parliament has called for new bans, such as that of automatic facial recognition systems in public places.
The Commission would like to authorize its use by law enforcement agencies in the fight against crime and terrorism. The subject should feed the debates with the Member States which refuse the prohibition of this controversial technology. The European Union hopes to conclude before the end of the year the first regulation in the world aimed at regulating and protecting innovation in artificial intelligence, a strategic sector in economic competition.
Brussels proposed an ambitious project two years ago, the examination of which has dragged on and which has been further delayed in recent months by controversies over the dangers of generative AI capable of creating texts or images. The European Parliament adopted its position on Wednesday in a plenary vote in Strasbourg. At the end of the day, negotiations must begin with the Member States to finalize the legislation as soon as possible. Commissioner Thierry Breton, who carried the text with his colleague Margrethe Vestager, called for the process to be concluded in “the coming months”.
“AI raises many questions – socially, ethically and economically. (…) It is a question of acting quickly and taking responsibility”, he declared on Wednesday. But the regulation will not come into force before 2026, in the best case. Believing that there was urgency, Thierry Breton and Margrethe Vestager announced their intention to obtain voluntary commitments from companies as quickly as possible.
Of great technical complexity, artificial intelligence systems fascinate as much as they worry. While they can save lives by enabling a quantum leap in medical diagnosis, they are also exploited by authoritarian regimes to exercise mass surveillance of citizens. The general public discovered their immense potential late last year with the release of California-based OpenAI’s editorial content generator ChatGPT, which can write original essays, poems or translations in seconds. Example of the feats now possible: an unreleased Beatles song recorded using AI to recreate the voice of John Lennon will be released this year.
But the dissemination on social networks of false images, more real than life, created from applications like Midjourney, has alerted to the risks of manipulation of opinion and the dangers for democracy. Scientists have called for a moratorium on the development of the most powerful systems, until they are better regulated by law.
“We need rules (…). The whole world is watching us,” said MEP Brando Benifei during a debate on Tuesday. Parliament’s position broadly confirms the Commission’s approach. The text is inspired by existing regulations on product safety and will impose checks based primarily on companies.
The heart of the project consists of a list of rules imposed only on applications deemed to be “high risk”. These would be systems used in sensitive areas such as critical infrastructure, education, human resources, law enforcement or migration management… Among the obligations: provide human control over the machine, drawing up technical documentation, or setting up a risk management system. Compliance with them will be monitored by supervisory authorities in each member country. The European Parliament intends to better take into account generative AIs of the ChatGPT type by calling for a specific regime of obligations which essentially repeat those provided for high-risk systems.
The Commission’s proposal, unveiled in April 2021, already provides a framework for AI systems that interact with humans. It will oblige them to inform the user that he is in contact with a machine and will force the applications generating images to specify that they were created artificially. An obligation that will probably be extended to texts. Bans will be rare. They will concern applications contrary to European values such as the citizen rating systems or mass surveillance used in China.
