UK Banks Still Lack Access to Anthropic's Mythos AI Model Amid Cybersecurity Concerns

Severity: Medium (Score: 43.0)

Sources: Letsdatascience, Itnews.Au

Published: 2026-05-31 · Updated: 2026-05-31

Keywords: banks, mythos, access, bank, still, model, anthropic

Summary

As of May 31, 2026, UK banks have not gained access to Anthropic PBC's Mythos AI model, which is crucial for testing defenses against cyber threats. Bank of England governor Andrew Bailey stated that the delay is due to political issues with the US administration. Despite Anthropic's willingness to provide trial access, the process has stalled. Bailey emphasized the importance of access given the potential cyber risks associated with advanced AI models. Some cybersecurity experts believe that fears of misuse of the model are exaggerated. The issue is compounded by US President Donald Trump's recent postponement of an executive order on AI, which would have facilitated developer engagement with the government. Bailey advocates for a global approach to managing cyber risks, highlighting the interconnectedness of financial institutions. The situation remains unresolved as banks continue to seek alternative models for red-teaming and penetration testing. Key Points: • UK banks lack access to the Mythos AI model for cybersecurity testing. • Political delays from the US administration are hindering access. • Experts suggest fears of the model's misuse may be overstated.

Detailed Analysis

**Impact** UK banks remain unable to access Anthropic's Mythos AI model six weeks after initial concerns, limiting their ability to test cyber defenses against advanced AI-enabled threats. The financial sector in the UK is directly affected, with potential operational risks due to delayed red-teaming capabilities. The Bank of England and the Financial Stability Board emphasize the need for a coordinated international approach given the interconnected nature of global banking systems. No specific data breaches or losses have been reported. **Technical Details** No detailed technical specifications, attack vectors, TTPs, malware, or CVEs related to Mythos have been disclosed. The model is described as advanced and powerful, with concerns focused on its potential misuse for cyberattacks rather than confirmed exploitation. The primary technical issue is restricted access for controlled testing and sandboxed environments to simulate AI-aided threat scenarios. No indicators of compromise (IOCs) are available. **Recommended Response** Defenders should monitor developments regarding controlled access programs or audit interfaces from Anthropic or similar AI providers to enable infrastructure red-teaming. Track US and international regulatory actions on pre-release engagement frameworks for advanced AI models. Maintain existing cybersecurity defenses and conduct threat modeling using alternative AI tools until Mythos access is granted. No specific patches or detections related to Mythos are currently available.

Source articles (2)

• UK banks still lack access to Mythos AI model — Itnews.Au · 2026-05-31
  British banks still have not been ⁠able to ⁠gain access to Anthropic's Mythos artificial intelligence model to check their systems against cyber threats, six weeks after it first drew concern, Bank of…
• Bank of England Says UK Banks Lack Mythos Access | Let's Data Science — Letsdatascience · 2026-05-31
  Bank of England governor Andrew Bailey told Bloomberg that British banks still have not gained access to Anthropic PBC's Mythos model six weeks after it first drew concern. Bailey said Anthropic was w…

Timeline

• 2026-04-15 — Concerns about Mythos AI model raised: The Mythos AI model drew significant concern regarding its potential misuse in cyber attacks.
• 2026-05-31 — Bailey comments on access delay: Bank of England governor Andrew Bailey stated that UK banks still lack access to the Mythos model due to political hold-ups.
• 2026-05-31 — Trump postpones AI executive order: US President Donald Trump postponed signing an executive order that would have established a framework for AI developer engagement.

Related entities

• Financial (Industry)