Anthropic has thrust itself into the centre of the latest debate over artificial intelligence after saying it created a model so powerful that it decided not to release it to the public. The company said the decision was based on a sense of responsibility and concerns about cybersecurity.
The model, called Mythos, quickly became a subject of discussion far beyond the AI industry. The announcement prompted reaction in Washington, where US treasury secretary Scott Bessent summoned the heads of major banks for a discussion about the model.
In the UK, the story also reached Westminster. Reform UK MP Danny Kruger wrote to the government urging it to “engage with AI firm Anthropic whose new frontier model Claude Mythos could present catastrophic cybersecurity risks to the UK”.
On social media, the response was immediate and intense. X, formerly Twitter, was soon filled with commentary as the company’s decision not to release the model drew attention from supporters, sceptics and critics alike.
Questions over motive
Anthropic says the model was withheld because it believed releasing it would be too risky from a cybersecurity standpoint. But that explanation has not satisfied everyone. Some sceptics argue that the announcement may have been intended to create a sense of urgency around the company’s technology, boosting its profile and helping it attract investment.
The dispute reflects a broader tension in the AI sector. Companies are under pressure to demonstrate that their systems are advanced enough to be commercially significant, while also reassuring governments and the public that those same systems are being handled responsibly.
Anthropic’s announcement plays directly into that tension. By framing Mythos as a model too dangerous to release, the company positioned itself as cautious and responsible. At the same time, the extraordinary nature of the claim helped amplify attention around its work.
That combination has made the episode a public relations test as much as a technical one. The central question is not only what Mythos can do, but also how much of the alarm around it reflects genuine concern and how much reflects the competitive realities of the AI industry.
For now, Anthropic has succeeded in putting its name at the centre of the conversation. Whether the reaction strengthens its credibility or fuels suspicion may depend on what more the company reveals about the model and the risks it says prompted the decision to keep it private.
