Ireland to investigate Google’s AI model over data protection concerns

Ireland’s Data Protection Commission (DPC) has initiated a cross-border investigation into Google Ireland Limited over data protection concerns.

The investigation focuses on the firm’s use of personal data in developing its Pathways Language Model 2. The commission announced its decision to look into Google’s PaLM2 AI system via a press release on September 12.

The entire aim of the inquiry is to find out whether Google complied with its obligations under the General Data Protection Regulation (GDPR). The GDPR is a mandatory obligation to follow for processing personal data of EU and EEA residents for AI model development purposes.

Ireland questions whether Google met DPIA requirement

The main motive of the investigation is to answer the question of whether Google conducted a required Data Protection Impact Assessment (DPIA) before processing personal data for PaLM 2.

In the press release, the DPC emphasized the importance of DPIAs in safeguarding individual rights and freedoms when data processing poses a high risk.

A spokesperson for the DPC said: “This statutory inquiry forms part of the wider efforts of the DPC, in regulating the processing of the personal data of EU/EEA data subjects in the development of AI models and systems.” 

Google introduced PaLM 2 in May 2023. It is an advanced language model said to have improved multilingual capabilities, reasoning skills, and coding abilities. The model is available in four sizes tailored to different use cases.

This investigation is part of a rising trend of increased scrutiny of tech companies’ AI development practices. Just a week prior, the DPC wrapped up an investigation into X. This resulted in the company agreeing to stop using personal data from EU and EEA users to train its AI chatbot, Grok.

X had also made a promise to delete the data it collected between May 7 and August 1. It also announced that it won’t gather any more data for the purpose of training Grok.