European Parliament blocks AI on lawmakers’ devices, citing security risks

The European Parliament has reportedly blocked lawmakers from utilizing the baked-in AI tools connected their enactment devices, citing cybersecurity and privateness risks with uploading confidential correspondence to the cloud.

Per an email seen by Politico, the parliament’s IT section said it could not warrant the information of the information uploaded to the servers of AI companies and that the afloat grade of what accusation is shared with AI companies is “still being assessed.”

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

Uploading information to AI chatbots, similar Anthropic’s Claude, Microsoft’s Copilot, and OpenAI’s ChatGPT, for example, means that U.S. authorities tin request the companies that tally the chatbots crook implicit accusation astir their users.

AI chatbots besides typically trust connected utilizing accusation that users supply oregon upload to amended their models, expanding the accidental that perchance delicate accusation uploaded by 1 idiosyncratic whitethorn beryllium shared and seen by different users.

Europe has immoderate of the strongest information extortion rules successful the world. But the European Commission, the enforcement assemblage that oversees the 27-member authorities bloc, past twelvemonth floated caller legislative proposals aimed astatine relaxing its information extortion rules to marque it easier for tech giants to bid their AI models connected Europeans’ data, drafting ire from critics who said the determination caves successful to U.S. exertion giants.

The determination to restrict European lawmakers from accessing AI products connected their devices comes arsenic respective EU subordinate countries reevaluate their relationships with U.S. tech giants, which stay taxable to U.S. instrumentality and the unpredictable whims and demands of the Trump administration.

In caller weeks, the U.S. Department of Homeland Security has sent hundreds of subpoenas demanding U.S. tech and societal media giants crook implicit accusation astir people, including Americans, who person been publically captious of the Trump administration’s policies. 

Google, Meta, and Reddit complied successful respective cases, adjacent though the subpoenas had not been issued by a justice and were not enforced by a court.