EU Privacy Regulator Opens Probe Into Google’s AI Model

The European Union (EU) privacy regulator opened an inquiry against tech giant Google on Thursday into whether the search engine adequately protected European Union users’ personal data before using it to help develop its foundational AI Model.

The probe concerns the Alphabet unit’s Pathways Language Model 2 (PaLM 2). The probe is being monitored by Ireland’s Data Protection Commission (DPC). The body is the lead EU regulator for most of the top US internet firms, due to the location of their EU operations in Ireland. 

“This statutory inquiry forms part of the wider efforts of the DPC, working in conjunction with its EU/EEA (European Economic Area) peer regulators, in regulating the processing of the personal data of EU/EEA data subjects in the development of AI models and systems,” the DPC said in a statement.

PaLM2 is a vast language model that supports various AI-powered services provided by Google, such as email summarization and other generative AI functions. These models bank on vast amounts of data to perform tasks, raising alarm about how they handle and treat personal information of citizens. 

The inquiry will scrutinize whether Google has breached its obligations under General Data Protection Regulation (GDPR) on the processing of the personal data of private citizens of the EU and European Economic Area. 

Under the GDPR framework, companies must conduct a data protection impact assessment before embarking on handling such information when there are chances that it might pose a threat to the rights and freedoms of individuals. This is particularly applied in new technologies and was “of crucial importance in ensuring that the fundamental rights and freedoms of individuals are adequately considered and protected”, the regulator said in a statement. 

This probe into PaLM2 is part of the ongoing trend in Europe, where regulators are increasingly screening the practices of major tech companies in terms of AI and data privacy. 

For instance,  Italy’s data privacy regulator temporarily banned ChatGPT in 2022 due to privacy violations, only allowing its return after its company OpenAI agreed to implement measures addressing the regulator’s concerns.

Also Read: Google Unveils 5 Exciting New Features For Android Devices

Earlier this month, Elon Musk’s social media platform X agreed to cease using user data to train its AI chatbot, Grok. This decision came after the DPC took legal action, filing an urgent High Court petition to prevent X from processing user data contained in public posts without consent.

Similarly in June, Meta held its plans to train its AI model Llama on public content shared by adults on Facebook and Instagram across Europe, following concerns raised by the Irish regulator. As a result, Meta limited the availability of some of its AI products to users in Europe. 

Google has declined to comment on the ongoing inquiry.