The Italian data protection authority (Garante per la protezione dei dati personali) has temporarily suspended the use of the artificial intelligence (AI) service ChatGPT in the country.
A privacy watchdog has launched an investigation into OpenAI’s chatbot and blocked its use due to allegations that it did not comply with Italy’s data collection regulations. Garante also claimed that OpenAI did not take sufficient measures to prevent users under the age of 13 from using his ChatGPT.
“We have noticed a lack of clear notification to users and all stakeholders whose data is collected by OpenAI, but above all, we believe it is important to justify the collection and large-scale storage of personal data to allow the algorithm to I realized that there is no legal basis for “training”. platform, ”reads the announcement (in Italian) announced today.
According to Tanium Chief Security Advisor Timothy Morris, the main problem in Italy seems to be ChatGPT’s anonymity.
“It comes down to a cost-benefit analysis. Most of the time, the benefits of new technology outweigh the drawbacks, but ChatGPT is a bit different,” Morris said. “Its ability to process extraordinary amounts of data and create easy-to-understand content that closely mimics human behavior is undeniably a game changer. It could lead to more regulation to oversee the industry.”
Additionally, Garante lamented that ChatGPT’s user data was not properly handled.
“It’s easy to forget that it’s only been a few weeks since ChatGPT has been in widespread use, and most users can’t stop thinking about the privacy implications of the data used to train the algorithms that underpin the product. commented Edward Machin. Senior Counsel at Ropes & Gray LLP.
“They may be happy to accept the deal, but the argument here is that users are not given the information to make an informed decision. […] We may not have a legal basis to process your data. “
In its announcement, the Italian privacy watchdog also mentioned a data breach that affected ChatGPT earlier this month.
More information on ChatGPT violations can be found here: Vulnerability in ChatGPT may have exposed users’ payment information
“AI and large language models like ChatGPT have great potential to be used for good and evil in cybersecurity. The emphasis will likely be on improving the capabilities of existing cybercriminals rather than activating armies of attackers.”
“Cybercrime is a multi-billion dollar organized criminal industry and ChatGPT is used to help smart criminals become smarter and stupid criminals more effective in phishing attacks. I plan to.”
OpenAI has until April 19th to respond to data protection authorities. If not, at most he could be fined €20 million or his 4% of annual turnover.The company has not yet responded to a request for comment Information security.
