Google’s AI model faces European Union scrutiny from privacy watchdog

European Union regulators stated Thursday they’re wanting into one in every of Google’s synthetic intelligence fashions over considerations about its compliance with the bloc’s strict information privateness guidelines.

Eire’s Information Safety Fee stated it has opened an inquiry into Google’s Pathways Language Mannequin 2, also called PaLM2. It is a part of wider efforts, together with by different nationwide watchdogs throughout the 27-nation bloc, to scrutinize how AI methods deal with private information.

Google’s European headquarters are based mostly in Dublin, so the Irish watchdog acts as the corporate’s lead regulator for the bloc’s privateness rulebook, often known as the General Information Safety Regulation, or GDPR.

The fee stated its inquiry is analyzing whether or not Google has assessed whether or not PaLM2’s data processing would doubtless lead to a “high risk to the rights and freedoms of individuals” within the EU.

Giant language fashions like PaLM2 are huge troves of information that act as building blocks for synthetic intelligence methods. Google makes use of PaLM2 to energy a spread of generative AI companies together with electronic mail summarizing. The corporate didn’t reply to a request for remark.

The Irish watchdog said earlier this month that Elon Musk’s social media platform X has agreed to completely cease processing consumer information for its AI chatbot Grok. The platform did so solely after the watchdog took it to courtroom the month earlier than, submitting an pressing Excessive Court docket utility to get X to “suspend, restrict or prohibit” processing of private information contained in public posts by its customers.

Meta Platforms paused its plans to make use of content material posted by European customers to coach the most recent model of its massive language mannequin after obvious strain from the Irish regulators. The choice “followed intensive engagement” between the 2, the watchdog stated in June.

Italy’s information privateness regulator final 12 months briefly banned ChatGPT due to information privateness breaches and demanded the chatbot’s maker OpenAI meet a set of calls for to resolve its considerations.

© 2024 The Related Press. All rights reserved. This materials is probably not printed, broadcast, rewritten or redistributed with out permission.