Italy lifts the ban on ChatGPT: what has OpenAI done to convince the regulator of the European country

The Italian Guarantor for the Protection of Personal Data (GPDP) has announced this Wednesday that ChatGPT is no longer blocked in the country. its residents can use the tool again of artificial intelligence with total normality, a change that takes place after OpenAI obeyed the demands of the Italian regulator.

The American company, for its part, has celebrated the measure. In a statement seen by AP has said: “We are excited to welcome you back [a los usuarios italianos] and we remain dedicated to protecting your privacy.” In this sense, OpenAI has ensured that it has “addressed or clarified” the problems and concerns raised.

ChatGPT works again in Italy

On March 31, the GPDP blocked ChatGPT throughout the Italian territory for the possible breach of the General Data Protection Regulation. At the time, the agency said it did not want to hinder the development of AI, but had taken the temporary measure to protect European privacy rules.

From the moment the release was published, a countdown was set in motion for OpenAI. The Microsoft-backed artificial intelligence company was required to communicate within 20 days the measures taken, otherwise it could face fines of up to 20 million euros (or 4 percent of annual global turnover).

The company’s CEO, Sam Altman, announced that it would collaborate with the Italian regulator, who later established a series of concrete measures that, if complied with, would allow the ban to be lifted and the return of the conversational chatbot to Italian territory. Most of them had an application deadline of April 30.

OpenAI did not sit idly by and implemented a series of changes to its application that, in some cases, went beyond Italian and even European territory. As explained by the GPDP, the application now provides “more transparency and rights to users”, which is why it has been rehabilitated.

The main changes adopted by OpenAI in ChatGPT

• The OpenAI website now shows detailed information about what personal data is collected and how it is processed for algorithm training.
• Users have the ability to object to their data being used to train AI models. There are two ways to do it, a new feature in the app and a dedicated form.
• You no longer need to register with the service to learn in detail about OpenAI’s data processing practices. This information is now available on the registration page.
• In Italy there is a new welcome page that refers to the new privacy policy and the methods of processing personal data for training AI models.
• The company has made it clear to users that while it will continue to process certain personal data to “ensure the proper functioning of the service”, it will process their personal data for the purpose of training algorithms, unless they indicate otherwise with the alternatives enabled.
• Already registered Italian users must declare that they are of legal age to use the service. Those over 13 years of age may also do so, but they must have the consent of their parents or legal guardians.
• New users will be required to provide their date of birth, and doing so will block registration if they are under 13 years of age and do not have parental or legal guardian consent to use the service.

It should be noted that privacy concerns surrounding ChatGPT were not raised solely by the Italian regulator. Other countries such as Germany and Spain have asked for explanations to the American company. In the case of the latter, he has launched a formal investigation. As the days go by we will see how this situation evolves within the European territory.

Images: Cover (Michele Bitetto +OpenAI)

In Xataka: OpenAI knows that ChatGPT is not perfect, so it will pay up to $20,000 to those who find bugs