The European Parliament announced the disabling of AI tools on devices designated for its members, in a step aimed at enhancing cybersecurity and protecting sensitive data.
According to an internal memo seen by European media outlets, the Parliament's IT administration indicated that it cannot guarantee the security of data sent to AI companies' servers. It also clarified that the volume of information that may be shared with these companies is still being assessed, considering that keeping these tools disabled is the safest option at present, especially given the possibility of confidential correspondence and documents being uploaded to cloud services outside the European Union, as reported by TechCrunch based on the memo.
Concerns About American Laws
Using AI-powered chatbots, such as Claude from "Anthropic", Copilot from "Microsoft", or ChatGPT from "OpenAI", effectively means sending data to companies subject to American law.
This raises the possibility of American authorities requesting user data to be handed over pursuant to legal orders, which opens the door to concerns about European data sovereignty.
Additionally, many of these tools rely on using input data to improve their models, which increases the likelihood of sensitive information being leaked or unintentionally reused.
Regulatory and Political Tension
The European Parliament's decision comes amid the European Union's adoption of one of the world's strictest data protection frameworks.
However, the European Commission had last year proposed legislative amendments aimed at facilitating the training of AI models on European data, which drew criticism from parties who saw this step as a concession to American technology companies.
This coincides with several member states reviewing their relationships with American tech giants, given these companies' continued subjection to American law and the ongoing changes in their internal policies.
American Companies Complying with Subpoenas
In recent weeks, the US Department of Homeland Security sent hundreds of subpoenas to technology companies and social media platforms, requesting data about individuals who had criticized US administration policies.
According to reports, companies such as Google, Meta, and Reddit complied in some cases, even in the absence of binding court orders.
In this context, the European Parliament's decision reflects a growing trend toward imposing strict restrictions on the use of AI tools within sensitive institutions, until the legal and regulatory frameworks governing cross-border data flows become clear. For all that's new, follow Arabic websites and specialized blogs, alongside what's offered by Egypt stores and Kuwait stores and vitamin stores and international websites, along with Mashhor for social media services.
