The Netherlands has no plans to block ChatGPT for now
The Netherlands is not yet going to block ChatGPT, a service that uses artificial intelligence (AI) to write text that appears to have been produced by humans. The Dutch Data Protection Authority (AP), the country's privacy regulator, said it is monitoring the situation closely.
The Italian privacy watchdog, the Garante per la Protezione dei Dati Personali (GPDP), decided on Friday to provisionally ban ChatGPT. According to the regulator, the company behind the system, OpenAI, does not comply with the rules for collecting personal data. Also, the platform does not have controls in place to check the age of minors using the system, and the company does not properly explain to users what personal data is collected and what happens to that data.
According to the GPDP, there was also a data breach at ChatGPT earlier this month. Data that was leaked about conversations that users had with the system and payment details of subscribers. In addition, the regulator complained that there is no legal basis at all for the massive collection and storage of data to "train" the algorithms.
OpenAI has 20 days to announce what measures are being taken to correct the flaws. If the company does not do this, a fine of up to 20 million euros, or 4 percent of annual turnover, is imminent.
The Dutch Data Protection Authority said that people sometimes share more personal information via apps and online services than they realize, or that they would like. "That is why we always warn you to be careful. For example, apps or services may seem free, but you actually 'pay' for them by agreeing in the terms and conditions to share all kinds of personal data. Sometimes even data outside the app, such as your internet behavior, your location data or your contacts."
The company behind ChatGPT, like all other companies in the European Union, must comply with European data protection rules. But the supervisors of the EU countries are responsible for enforcement, said the European Commission. The executive body of the EU may release a more extensive response later, a spokesperson said. The Commission has been working on AI regulations for some time now. However, due to the rapid developments in the field of AI, the regulations again need to be modified. As a result, it may take longer before the EU countries and the European Parliament can consider it.
The emergence of this kind of "generative" artificial intelligence also worries critics. One European Union agency warned against abuse of the program in fraud attempts, spreading untruths and cybercrime. There are also ethical and legal objections to the use of artificial intelligence, which is becoming increasingly sophisticated.
An estimated 100 million people worldwide use ChatGPT every month. That program caused a worldwide sensation because, based on a few simple instructions, it was able to create complete text material that can be indistinguishable from human-produced writing. The program does this by looking for patterns in huge amounts of data.
Reporting by ANP