He European Parliament on Wednesday opposed real-time biometric surveillance and imposed conditions for the new systems, such as ChatGPTby approving the artificial intelligence law that is being negotiated in Brussels to promote its development, while limiting the possible risks.
With 499 votes in favour, 28 against and 93 abstentions, the European Parliament thus established its position for the negotiations that will begin today with the European Commission and the Council of the EU to agree on the final text of the law and that will continue under the rotating presidency that Spain will assume on July 1.
Among the demands of European Parliament There is the prohibition of biometric surveillance in real time in public spaces, contrary to what the Community Executive and the EU countries have defended, which are committed to allowing it if authorized by a judge, in very specific cases.
GPT
In full debate on the need to regulate the systems that are capable of generating text and audiovisual material, Parliament wants its developers to demonstrate that they have reduced the risks that technology can generate to health, fundamental rights and the democratic system.
MEPs also want users who use this technology to generate false images (‘deep fakes’) to specify that the content has been manipulated by artificial intelligence.
They also ask that the developers of these systems explain in a “sufficiently detailed” way what copyright-protected data they have used to train artificial intelligence.
biometric identification
The European Parliament opposed the use of real-time biometric identification systems in public spaces. The MEPs, however, did open the door to being able to use them a posteriori, always with judicial authorization, to prosecute “serious crimes”.
The European Parliament thus rejected the amendment of the European People’s Party, more lax with the protection of fundamental rights.
The PPE was in favor of being able to use these systems, with the permission of a judge, to search for missing persons (including minors), to prevent a terrorist attack and to locate those who have committed a crime that carries at least three years of imprisonment. prison.
“If there is a terrorist attack on the street right now or if the police have received a notice that a child has been lost, images can already be obtained from surveillance cameras, in accordance with current laws and with judicial authorization, biometric identification can be used”, recalled one of the law’s rapporteurs, the Liberal MEP Dragos Tudorache.
High risk
The regulation classifies a whole series of security systems as high risk. artificial intelligence with very specific uses that can only be introduced to the market if they respect the fundamental rights and values of the EU.
For example, those that can be used to influence the outcome of an election, those used by financial institutions to assess the solvency and establish a person’s credit rating or those used by the border control police to control, monitor or process data to predict migratory movements.
After a vote that the Vice President of the European Commission for the Digital Age, Margrethe Vestager, described as “historic” as it is a pioneering standard in the world, negotiations will begin between the three European institutions to agree on the final text of the law.
In case the talks were to conclude this year, the rule would not begin to be applied until 2026 -although the intention is to be able to reduce the deadlines- so Brussels has proposed an agreement between the big technology companies so that they immediately label all the false content that is generated with artificial intelligence.
The general director of the Association of European Consumers (BEUC), Ursula Pachl, assured that “the prohibitions that the European Parliament has proposed today on facial recognition in public spaces or special categorization by companies are essential to protect fundamental rights ”.
MORE NEWS:
EFE International news agency based in Madrid and present in more than 110 countries.