Grok в тисках закона: почему Британия ополчилась на ИИ Маска

Elon Musk has always loved playing on the edge of the rules, but this time his penchant for provocation met serious resistance. While the billionaire promotes his AI chatbot Grok as the only "honest" alternative to the polished neural networks from Google and OpenAI, British authorities decided to check whether this honesty crosses the line of law. Two major UK regulators announced the start of official investigations into xAI and the X platform. This is not just a routine inspection, but a full-fledged legal challenge that could change the rules of the game for all AI developers in Europe.

The first blow was struck by the British Information Commissioner's Office. The regulator suspects that companies X Internet Unlimited (X) and xAI violated personal data processing rules. The problem is as old as the world: millions of posts from X social network users were used to train Grok, and no one asked for their consent. Musk simply switched on the "train on my data" toggle by default, leaving users only the ability to manually dig through settings to turn it off. In a world of strict GDPR rules and British data protection legislation, such tricks usually end in multi-million-pound fines.

The second front was opened by Ofcom, which oversees the safety of online platforms. Here the claims are even more serious. The regulator is studying Grok's ability to generate harmful content, including pornographic deepfakes and sexually explicit images. Musk positioned his AI as a system without "progressive filters," but it turned out that the absence of censorship easily turns the neural network into a tool for creating toxic garbage. For British authorities, who recently passed the Online Safety Act, this is a direct challenge to their authority.

The situation with Grok is a classic clash of two worlds. On one hand — Silicon Valley with its principle of "move fast and break things," where Musk is trying to create an AI rebel. On the other — European bureaucracy, which is no longer willing to turn a blind eye to how technology giants use citizen data for their experiments. Musk has already clashed with European officials, calling them censors, but the legal consequences of the investigation could be far more painful than verbal battles on social networks.

If the ICO proves that British data was used illegally, xAI could face demands to delete all progress made using this data. This effectively means a reset of the model's progress for this specific region. And Ofcom could force the platform to implement such strict filters that none of Grok's former "freedom" would remain.

For Musk, this is a stalemate: either bend to the regulators and become "like everyone else," or risk complete blockage of his AI products in one of the key markets. The key point: Britain is setting a precedent where AI is punished not for mistakes, but for the very ideology of "absence of filters." Will Musk be able to defend his project, or will Grok become another victim of European regulation?