The Italian data protection authority decided this Friday to temporarily block the artificial intelligence application ChatGPT for not respecting user data.
The decision, “effective immediately”, results in the “temporary restriction of the Italian user’s data processing in relation to” the OpenAI application, the Italian regulator said in a statement.
The suspension is temporary until the artificial intelligence company complies with “the regulations” on data protection in force in the European Union, says the Italian authority, which has opened an investigation into ChatGPT.
The regulator said there was no legal basis to “justify the collection and mass storage of personal data for the purpose of ‘training’ the algorithms underlying the operation of the platform”.
He said OpenAI, for example, does not verify users’ age and “exposes minors to absolutely inappropriate responses compared to their level of development and self-awareness.”
The American company now has 20 days to respond to the concerns of the Italian regulator, otherwise a fine of 20 million euros will follow or about 4% of annual turnover will be withheld.
Based on the GPT-3.5 model (Generative Pre-trained Transformer version 3.5), ChatGPT works as a conversational bot where you can ask questions and get answers in seconds. After creating an account, the interface is simple: there’s a box to ask questions and the answer appears as if someone was typing it in real time on the other end.
It is possible to have conversations in Portuguese from Portugal because the prototype is multilingual and adapts to the user. That is, if someone starts a conversation in English and then switches to Portuguese, ChatGPT adapts and starts responding in Portuguese.
Source: DN
