KEY TAKEAWAYS
- British banks have still not been able to gain access to Anthropic’s Mythos model to check their systems against cyber threats, six weeks after it first drew concern.
- Bailey said Anthropic is willing to share the model on a trial basis, but the delay appears to be “somewhat caught up in the process with the US administration.”
- UK banks are currently using alternative AI models to test their cyber defences in lieu of Mythos access.
- Bailey, who also chairs the Financial Stability Board, stressed that cyber risk spillovers are “so big that we can’t just have a single sort of national approach.”
Bank of England Governor Andrew Bailey confirmed on Friday that British banks still have not been granted access to Anthropic’s Mythos artificial intelligence model, six weeks after it first emerged as a major concern for the UK financial sector.
Speaking on the sidelines of a central banking conference in Reykjavik, Bailey told Bloomberg Television that Anthropic had expressed willingness to share Mythos on a trial basis, but a political blockage in Washington was preventing that from happening.
The disclosure arrives as the UK’s financial regulators and frontier AI safety warning remains one of the most active live issues in British technology governance.
The Political Hold-Up Bailey Cannot Fully Explain
Bailey said: “It hasn’t happened yet and I think this has been somewhat caught up in the process with the US administration. Quite why the process differs from one company to another, I’m afraid I can’t explain to you.”
The comment points to ongoing tensions between Anthropic and the Trump administration, which designated the company a “supply-chain risk” earlier this year after Anthropic refused to grant the Pentagon unrestricted access to its models.
That dispute, now being contested in federal court, appears to be creating procedural friction around government-adjacent Mythos access, even when requests come from allied central banks rather than US federal agencies.
Last month, Bailey said Anthropic “may have found a way to crack the whole cyber risk world open,” a notable endorsement from a central bank governor that highlighted how seriously the BoE viewed Mythos’s offensive capabilities.
Six weeks later, the access he sought has still not materialised.
Why Bailey Says a Global Approach Is Non-Negotiable
Bailey further added: “Spillovers from this sort of cyber risk are so big that we can’t just have a single sort of national approach. Anybody who thought, ‘Well, I’ve dealt with my banks, that’s okay’, I’m afraid that won’t work, because they’re all so heavily interconnected.”
The Financial Stability Board, which Bailey chairs, is the international body best positioned to establish a coordinated access framework.
However, its decisions require consensus across member jurisdictions, and the US political dimension currently prevents that consensus from forming smoothly. In the meantime, Bailey confirmed that UK banks are using other AI models to test cyber resilience.
Some cybersecurity experts have since told Reuters that fears of unrestricted hacking with Mythos are overstated, but Bailey’s public statements suggest the BoE’s concern has not diminished and that access remains a priority for British financial stability regulators.
The development comes alongside the NatWest and Amazon venture banking unit, itself a sign of how deeply UK financial institutions are integrating AI infrastructure, making the Mythos exclusion increasingly conspicuous.
Source: BOE’s Bailey Says UK Banks Still Lack Access to Mythos
