The European Parliament has found an unconventional approach to the problem of uncontrolled AI use by its members. Instead of total bans, the institution is launching its own platform, EPGenAI Hub, providing access to models from OpenAI, Meta, Anthropic, and France's Mistral. The goal is to steer politicians away from public websites toward a secure alternative.

Anxiety within the EU is growing daily. AI-generated texts are infiltrating parliamentary work too quickly. Members of Parliament and their assistants are increasingly using services to write speeches, inquiries, and amendments to laws. Approximately 2,100 people—nearly a fifth of the staff—use neural networks daily.

At a July meeting of several committee chairs, alarm bells were raised. Suspicion was reinforced by an abnormal surge in amendments requiring translation. The texts featured a characteristically dry style and contained references to non-existent laws.

Brando Benifei, who oversaw work on the AI Act, acknowledged that the parliament is facing a wave of "neural slop" and fakes. Irish MP Barry Andrews, who chairs the development committee, noted that some colleagues are writing inquiries and speeches using AI. According to him, this is a personal choice due to the enormous workload, but for Andrews himself, such tasks remain key responsibilities of a parliamentarian. For now, lawmakers are not required to disclose their use of neural networks.

The new platform has already been presented to MPs and staff in workshops. The official launch of the project is scheduled for September. The beta version allows users to choose between local models (Llama from Meta, GPT-OSS from OpenAI, Mistral Small) and external services (ChatGPT and Claude Sonnet).

However, this is where the main paradox lies. By launching the platform, the parliament is expanding the presence of American AI giants—amid Brussels' push to reduce technological dependence on the US. Previously, the institution replaced Google with the French search engine Qwant. Additionally, technicians disabled the built-in AI on MPs' tablets for cybersecurity reasons.

The parliament's press service clarified an important detail: the list of available models may change in the coming months. Sovereign European solutions from new cloud providers are planned to be added to the system.

Similar stories are unfolding in Russia. MPs have been accused of using ChatGPT to draft laws. Journalists suspected AI involvement in the preparation of bill No. 1126815-8. The reason was a footnote in the explanatory note: a link to a study by the "Pulse of NCO" project contained the tag source=chatgpt.com. Such traces typically remain when copying from the chat.

Representatives of the State Duma explained the situation: a staff member used AI only for quickly finding the original source, and the data itself turned out to be completely accurate. The committee added that drafting bills requires deep knowledge, and neural networks are not yet capable of handling such tasks.

Expert commentary: The story of EPGenAI Hub is a vivid example of how regulators are trying to harness a technology they cannot control. Instead of bans—creating a "safe" AI environment, but with the same American models. This is not so much a fight against neural slop as an attempt to manage the inevitable. The question is how effectively the parliament will be able to filter its own content when one in five employees already uses AI daily.