Italy’s data protection authority has lifted the nationwide ban on OpenAI’s ChatGPT. On Friday, April 28, the Garante Per La Protezione Dei Dati Personali (GPDP) confirmed that OpenAI has taken steps to address the concerns laid out in its order, which temporarily restricted access to the chatbot in the country.
Among other things, OpenAI’s website now has an age verification tool to check the ages of Italian users before they sign up. Existing users will see a welcome back page with a button to confirm their age.
In a press release on March 31 announcing the ban on ChatGPT, Italy’s privacy watchdog accused OpenAI of failing to verify the ages of its users. The GPDP also said there’s “no legal basis underpinning the massive collection and processing of personal data in order to “train” the algorithms on which the platform relies.”
Italy’s ChatGPT ban led privacy regulators in other countries, like Germany, Italy, and France, to take a closer look at the chatbot. The European Data Protection Board (EDPB) also announced a probe into ChatGPT for suspected GDPR violations.
“The Italian SA welcomes the measures OpenAI implemented and calls upon the company to comply with the additional requests laid down in its 11 April order,” the GPDP said.
Personal Data Opt-Out Form, New Privacy Controls
Along with the new age verification tool, individuals in Europe can now choose not to allow OpenAI to use their data to train its algorithm. Users who want the company to delete their data can put in a request by filling out a form. The form requires users to provide some personal details, like their names, along with the relevant prompts that show evidence of data processing.
“Under certain privacy or data protection laws, such as the GDPR, you may have the right to object to the processing of your personal data by OpenAI’s models,” the form reads.
OpenAI recently added added “incognito mode” to ChatGPT, allowing users to disable their chat histories and chatbot training.
“While history is disabled, new conversations won’t be used to train and improve our models, and won’t appear in the history sidebar,” OpenAI’s website reads.
As OpenAI works to resolve privacy concerns surrounding its chatbot, cybersecurity experts have warned that ChatGPT poses a security risk.
Despite efforts to make ChatGPT-4 more resistant to prompts to write malicious code, experts at the recent RSA Conference 2023 showed that cybercriminals can get the chatbot to generate ransomware code using a workaround. It can also be used to craft convincing phishing emails.
The EU’s AI Act, published two years ago, hardly took advanced chatbots into account, referencing only deepfakes as AI-generated material that requires regulation. The AI Act has since been updated after the release of ChatGPT. On Thursday, April 27, lawmakers revealed new draft legislation focusing on copyright protection as a key part of regulating AI.
How to Block ChatGPT From Collecting Your Data
If you’re concerned about your privacy and don’t want OpenAI to continue using your data to train its chatbot, go to ChatGPT > Data Controls to disable chat history and model training. If you’re in the EU, you can also get OpenAI to delete your data by filling out this form.
While ChatGPT is now available in Italy, it’s still inaccessible in other countries like China, Russia, Iran, and North Korea. If ChatGPT is blocked in your country, our guide to accessing ChatGPT explains how to circumvent geo-based restrictions and access the chatbot.
