The Dutch Data Protection Authority has issued a warning to voters ahead of next week's general elections, urging them not to seek voting advice from artificial intelligence-based chatbots. A study conducted by the authority showed that these chatbots provide a polarized and distorted view of the Dutch political landscape. Monique Verdier, the vice president of the institution, emphasized the lack of transparency and verifiability in how these systems operate. In the framework of the experiment, the authority tested four chatbots, including ChatGPT and Gemini, observing that left-wing voters were directed towards the GroenLinks-Labour party, while right-wing voters were guided towards the PVV, the far-right party. The authority highlighted that these chatbots could be classified as high-risk systems under the EU AI Act, which will impose additional rules starting next year.
Sources
Nu lăsați chatboturile AI să vă spună cum să votați, avertizează autoritățile olandeze alegătorii
Alegătorii olandezi au fost avertizați. De ce nu ar trebui să ceară sfaturi electorale inteligenței artificiale
Alegeri Olanda – Autorităţile avertizează cetăţenii să nu întrebe inteligenţa artificială cu cine să voteze
Latest News
20:00
A Romanian woman and four relatives are on trial over the death of a baby abandoned in Austria
19:54
Dan Reşitnec was dismissed from the position of Secretary General of the Government
19:25
Italian authorities dismantled a network that exploited 84 foreign workers, including Romanians, confiscating nearly 800,000 euros
19:19
A celebration of the Romanian community at the center of a political dispute in Italy / Criticism over the absence of the opposition and accusations of the event’s political use
19:10
EP rapporteur calls for €47.4 billion for Erasmus+ in 2028–2034 and grants adjusted to the cost of living
See more news