UPDATE: OpenAI says it's stopped offering ChatGPT to users in Italy to comply with the country's investigation into ChatGPT. However, company CEO Sam Altman tweeted: "Though we think we are following all privacy laws."
Original story:
An Italian regulator is ordering a ban on ChatGPT for violating the EU’s privacy rules through the “unlawful” collection of user data.
The Italian Data Protection Authority today announced it had “imposed an immediate temporary limitation” that bars OpenAI from processing users’ data in the country.
ChatGPT’s developer, OpenAI, has a privacy policy in place. However, the Italian regulator says ChatGPT is not complying with Europe’s GDPR privacy law, which requires companies collecting and storing data on EU citizens to properly disclose how they’re doing it.
“More importantly, there appears to be no legal basis underpinning the massive collection and processing of personal data in order to ‘train’ the algorithms on which the platform relies,” the agency says.
Italy’s data protection authority also faults OpenAI for neglecting to add an age-verification system for those aged 13-18. It also notes that ChatGPT can make mistakes and produce misinformation about users. On top of all this, the regulator cites ChatGPT for accidentally leaking users’ private information last week as another cause for concern.
The Italian regulator has launched an investigation into OpenAI. If the company fails to comply with the privacy demands within 20 days, then OpenAI could face a fine of up to €20 million or 4% of the company’s annual revenue, the government agency warned.
OpenAI didn’t immediately respond to a request for comment. But it looks like ChatGPT remains available in the country. On Friday morning, PCMag was still able to access the AI chatbot via a VPN server based in Italy.
Still, the order may signal that the European Union as a whole could go after ChatGPT for violating GDPR. It also comes as some tech entrepreneurs and computer scientists have called on the tech industry to hit pause on AI research until society and elected leaders can weigh the potential consequences. One nonprofit has also urged the US Federal Trade Commission to investigate OpenAI for allegedly failing to place enough safeguards around ChatGPT.