ROME - In the aftermath of a data breach, Italy is temporarily suspending the artificial intelligence program ChatGPT as it investigates a suspected infringement of rigorous European Union data protection standards, the government's privacy authority said Friday.
The Italian Data Protection Authority stated that it was implementing interim measures "until ChatGPT respects privacy," including temporarily prohibiting the firm from processing data from Italian users.
OpenAI, the company that created the chatbot, announced late Friday night that it had disabled ChatGPT for Italian users at the government's request. The firm stated that it thinks its processes are in accordance with European privacy rules and that it wants to make ChatGPT available again shortly.
While some public schools and universities around the world have blocked ChatGPT from their local networks due to student plagiarism concerns, Italy's action is "the first nation-scale restriction of a mainstream AI platform by a democracy," according to Alp Toker, director of the internet access monitoring organization NetBlocks.
The limitation applies to the web version of ChatGPT, which is popular as a writing helper, but it is unlikely to disrupt software applications from firms that already have licenses with OpenAI to utilize the same technology that powers the chatbot, such as Microsoft's Bing search engine.
Large language models, which enable such chatbots, are AI systems that can replicate human writing styles based on a massive repository of digital books and online writings that they have digested.
The Italian authority stated that OpenAI must report within 20 days on the steps it has taken to safeguard the privacy of its customers' data or risk a punishment of up to 20 million euros (almost $22 million) or 4% of annual worldwide sales.
The statement mentions the EU's General Data Protection Regulation and refers to a recent data breach affecting ChatGPT "users' discussions" and subscriber payment information.
OpenAI previously reported that ChatGPT would be taken offline on March 20 to address an issue that permitted certain users to read the titles, or subject lines, of other users' conversation histories.
Sam Altman, CEO of San Francisco-based OpenAI, said this week that he will travel on a six-continent journey in May to discuss the technology with consumers and developers. This includes a stop in Brussels, where European Union legislators are drafting new laws to limit high-risk AI capabilities, as well as stops in Madrid, Munich, London, and Paris.
BEUC, the European Consumer Organization, demanded on Thursday that EU authorities and the bloc's 27 member countries examine ChatGPT and similar AI chatbots. According to BEUC, it might be years until the EU's AI law takes effect, therefore authorities must act quickly to safeguard consumers from potential threats.
“In only a few months, we have seen a massive take-up of ChatGPT, and this is only the beginning,” Deputy Director General Ursula Pachl said.
Waiting for the EU's AI Act is "insufficient" since "real worries are arising about how ChatGPT and similar chatbots may deceive and manipulate people."
