European Parliament approves project to regulate AI – 06/14/2023 – Tech

The European Parliament approved this Wednesday (14) a project to regulate the use of AI (Artificial Intelligence) in the EU (European Union), which means the beginning of a delicate negotiation with representatives of the 27 countries of the bloc.

The approved rule (with 499 votes in favor, 28 against and 93 abstentions) will regulate AI according to the level of risk: the higher it is considered for rights, or for people’s health, for example, the greater the obligations of systems technological.

The high risk list includes this technology in critical infrastructure, education, human resources, public safety and migration management.

The European Parliament has included additional conditions for the high-risk classification to be established, including the possibility of harming health, safety, fundamental rights, or the environment.

With the conditions, parliamentarians included in the package a ban on the widespread use of automatic facial recognition systems in public places.

Under the approved rule, the EU will be able to authorize the use of AI devices for recognition as a support in the fight against crime, but the specific details must still be negotiated with the countries of the bloc.

It also defines special requirements for generative AI systems, such as ChatGPT and DALL-E, capable of producing text, images, code, audio and other resources, and includes the obligation to inform users that a machine, not a human being, produced the content.

lengthy negotiations

The EU’s original plans to regulate the use of AI are for 2021, but the topic gained urgency after ChatGPT entered the scene and highlighted the rapid development of the sector.

Now, direct negotiations will begin with the 27 countries of the bloc, with the goal of reaching an agreement by the end of the year.

Even if the goal is achieved, the standard should only come into force in 2026 and therefore the EU will also need to negotiate an interim and voluntary framework with technology companies that develop AI.

The European Commissioner for the Internal Market, Thierry Breton, appealed to the countries of the bloc and to the European Parliament for the negotiations to be concluded in the “next months”.

The use of AI, highlighted Breton on Tuesday (13), raises many questions, on the social, ethical and economic levels. It is about acting quickly and that we all assume our responsibilities.

One of the lawmakers who supported the initiative, Brando Benifei, called for a “common approach” to managing AI risks. “We need to compare notes with lawmakers around the world,” he said.

In addition, lawmakers included in the bill some provisions on banning AI systems for biometric surveillance, emotion recognition and predictive surveillance.

The dissemination of false images on social networks, created with AI-based applications such as Midjourney, was a warning sign about the risks of manipulation of public opinion and the dangers for democracy.

For this reason, European scientists have called for a moratorium on the development of the most powerful systems, until more efficient regulation by law.