Italy recently made headlines by becoming the first Western nation to ban the now infamous advanced chatbot ChatGPT.
This comes on the heels of many tech titans recently signing a letter asking for a six-month pause on additional artificial intelligence (AI) development — a pause that the head of ChatGPT maker OpenAI has so far refused to sign on to. Against the backdrop of mounting concerns about AI running amok, Italy has now taken definitive action.
Italian authorities banned the use of ChatGPT saying they believe it violates general data protection regulation (GDPR), which regulates how private and public entities can gather, store and process personal data. One of the main underpinnings of ChatGPT’s technology is that it collects data from users to help it “learn” and become more effective at answering user inquiries.
The country has since allowed the AI chatbot to return if they take “useful steps” to mitigate some of these concerns.
This type of mass data collection is of particular concern to Italian regulators who believe that ChatGPT doesn’t have the legal right to “the mass collection and storage of personal data for the purpose of ‘training’ the algorithms underlying the operation of the platform.” But it’s this way of collecting data that powers ChatGPT’s ability to process information and answer questions intelligently.
To stay updated with top startup news & investments, sign up for Benzinga’s Startup Investing & Equity Crowdfunding Newsletter
Italian regulators also voiced concern that ChatGPT potentially “exposes minors to absolutely unsuitable answers compared to their degree of development and awareness.” It’s not a stretch of the imagination to say that Italy’s ban goes to the heart of how ChatGPT works. During the ban period, Italian privacy regulators will look into how ChatGPT functions. If they find the underlying processes to be in violation of Italian law, the ban could become permanent.
Alphabet Inc.’s AI-powered chatbot, Bard has escaped the ban so far because it’s only available to users older than 18, a feature implemented to alleviate regulators’ privacy concerns. Italy’s data regulator gave OpenAI 20 days to respond to its privacy concerns. If OpenAI fails to respond in that time, it could face a fine of nearly $22 million.
Italy is not alone in its concerns about the inner workings of ChatGPT. Regulators in Ireland and the U.K. are both keeping a close eye on the situation and may respond with their own bans, investigations and/or fines. OpenAI believes it is in compliance with existing regulations and has responded by disabling ChatGPT for users in Italy.
ChatGPT has made headlines lately for its ability to interpret information and answer questions in real-time. Aside from its potential benefits, this has led to some legitimate concerns about the technology behind it being misused by everyone from freelance writers to college students looking for a quick, easy way to turn out quality information in a way that was unimaginable as recently as 10 years ago.
Italy’s action comes against a backdrop of legitimate fears about the damage AI-powered apps like ChatGPT could do if the people who design them lose control. ChatGPT is already banned by more repressive regimes such as China, North Korea, Russia and Iran.
Both the EU and the United States have privacy regulators working overtime to stay ahead of the curve. This opening move by Italy to ban ChatGPT is the beginning of an inevitable, high-stakes chess match between AI developers and regulators. The results will have massive implications for AI and AI startups moving forward.
See more on startup investing from Benzinga.