OpenAI’s ChatGPT, which took the world by storm after its launch in November, now faces a temporary ban in Italy over “unlawful” collection of personal data and the lack of an age verification system for children.
Italy’s data protection authority, known as Garante, was prompted to act by ChatGPT’s data breach on March 20, which OpenAI says allowed a number of users to see other users’ information, such as their first and last names and email addresses , and the last four digits of your credit card number.
Garante accused the US-based AI company of having “no legal basis” justifying “the massive collection and processing of personal data” needed to “train the algorithms that the platform relies on.” It also added that the lack of an age verification mechanism exposes children to “inappropriate” responses.
As a result, the Italian regulator on Friday launched an investigation into OpenAI and temporarily blocked access to GhatGPT in the country.
Visit us at the TNW conference on June 15th and 16th in Amsterdam
Get a 20% discount on your ticket now! Limited time offer.
In response, Sam Altam, CEO of OpenAI, tweeted that GhatGPT had suspended its services in Italy, but stated that he didn’t think the company had violated privacy laws.
We are of course bowing to the Italian government and have shut down ChatGPT in Italy (although we believe we comply with all data protection laws).
Italy is one of my favorite countries and I look forward to visiting again soon!
— Sam Altman (@sama) March 31, 2023
Open AI has 16 days to respond with adequate security measures or risk either a fine of up to €20 million or 4% of its total worldwide annual turnover.
ChatGPT’s disturbing capabilities are raising concerns about the exponential progress of Generative AI.
At the end of March, the European Consumers Organization (BEUC) called on EU and national authorities to investigate OpenAI’s system.
“Despite all the benefits that AI can bring to our society, we are currently not adequately protected from the harm it can do to people,” Ursula Pachl, deputy director general of BEUC, said in a statement.
The organization fears that until the union’s AI law comes into force, consumers risk harm from under-regulated technology they are unprepared for.
Across the Atlantic, an open letter signed by AI experts and industry leaders called on AI labs to immediately suspend training of systems more powerful than GPT-4, ChatGPT’s successor, for at least six months.
The signatories – including DeepMind researchers, computer scientist Yoshua Bengio and Elon Musk – stressed the need for regulatory policy, stressing that “powerful AI systems should only be developed when we are confident that their impacts will be positive and their risks manageable are .”
Resistance from the tech industry
The Italian regulator’s move to suspend access to ChatGPT could result in a loss of economic opportunity in the country, Daniele Servadei, founder of Bologna-based SaaS e-commerce startup Sellix, told TNW.
“It restricts the development and growth of the AI and technology sector and prevents local companies from using the technology,” he said.
Servadei further noted that the ban could have “a chilling effect” on AI development in Italy and beyond, “as developers and investors may fear future bans or restrictions on their products, which could hamper innovation and investment in the industry.”
Simone Basso, chief product and technology officer at Milan-based travel scaleup WeRoad, also highlighted the potential negative impact on local businesses.
“ChatGPT has boosted the productivity of teams at WeRoad and around the world, but recent decisions in Italy – from banning synthetic meat research to holding back 5G rollouts – have hampered the country’s technology, innovation and productivity growth,” Basson told TNW.
Basson added that while consumers can bypass Garante’s restrictions with a VPN, companies with solutions based on such technologies “face uncertainty” and will be forced to prioritize other markets.
Is Italy setting a precedent for Europe?
While ChatGPT was already unavailable in mainland China, Hong Kong, Iran, Russia and parts of Africa, Italy is the first Western country to crack down on the AI system.
Germany could soon follow Italy, said German data protection commissioner Ulrich Kelber Handelsblatt.
Authorities in France and Ireland are now in contact with the Italian regulator, Reuters reports.
“We are following this with the Italian regulator,” said a spokesman for Ireland’s data protection commissioner. “We will coordinate with all EU data protection authorities on this matter.”
Regulators’ concerns about ChatGPT – coupled with its popularity as the fastest growing consumer app in history – highlight the need for coordinated and transparent EU policies regulating the use and development of large language models.
It’s still unclear how ChatGPT and similar models will be affected by the EU’s upcoming AI law, which aims to strike a difficult balance between security and innovation.