European Parliament blocks AI on lawmakers devices, citing security risks | TechCrunch

The European Parliament has reportedly blocked lawmakers from using the baked-in AI tools on their work devices, citing cybersecurity and privacy risks with uploading confidential correspondence to the cloud.

Per an email seen by Politico, the parliament’s IT department said it could not guarantee the security of the data uploaded to the servers of AI companies and that the full extent of what information is shared with AI companies is “still being assessed.”

As such, the email said, “It is considered safer to keep such features disabled.”

Uploading data to AI chatbots, like Anthropic’s Claude, Microsoft’s Copilot, and OpenAI’s ChatGPT, for example, means that U.S. authorities can demand the companies that run the chatbots turn over information about their users.

AI chatbots also typically rely on using information that users provide or upload to improve their models, increasing the chance that potentially sensitive information uploaded by one person may be shared and seen by other users.

Europe has some of the strongest data protection rules in the world. But the European Commission, the executive body that oversees the 27-member state bloc, last year floated new legislative proposals aimed at relaxing its data protection rules to make it easier for tech giants to train their AI models on Europeans’ data, drawing ire from critics who said the move caves in to U.S. technology giants.

The move to restrict European lawmakers from accessing AI products on their devices comes as several EU member countries reevaluate their relationships with U.S. tech giants, which remain subject to U.S. law and the unpredictable whims and demands of the Trump administration.

In recent weeks, the U.S. Department of Homeland Security has sent hundreds of subpoenas demanding U.S. tech and social media giants turn over information about people, including Americans, who have been publicly critical of the Trump administration’s policies.

Google, Meta, and Reddit complied in several cases, even though the subpoenas had not been issued by a judge and were not enforced by a court.