UK banks blocked from cyber AI tool Mythos get offer from rival OpenAI
1 hour ago
Zoe Kleinman Technology editor
Reuters
OpenAI has offered nine major UK banks access to its cyber security AI tool GPT-5.5 Cyber, as its fierce rival Anthropic has blocked them in previews of its version, Claude Mythos.
Both tools are designed to find hidden security weaknesses in digital systems and can outperform humans at some hacking and cyber-security tasks.
Mythos caused a stir when it was announced in April, with Anthropic claiming it had found a weakness in one legacy system which had remained hidden for nearly 30 years.
Some finance ministers, central bankers and financiers have since expressed serious concerns about it, fearing the model could undermine the security of all financial systems.
PA Media
Governor of the Bank of England Andrew Bailey said last week UK banks still could not access Mythos, Anthropic's AI model
The AI Security Institute, which has tested both Anthropic's Mythos and OpenAI's GPT-5.5 Cyber, said in its report that they reached "a similar level of performance" in the tasks it set them.
Last week Bank of England governor Andrew Bailey warned that UK banks were still unable to access Mythos to test the security of their own digital systems and apps.
Former UK Chancellor George Osborne, who is now a senior executive at OpenAI, told the BBC that Bailey had not contacted him directly. He said his firm did not want to "hide [5.5 Cyber] away or keep it to ourselves" but confirmed that, like Mythos, it would not become available to all.
"The key things with these tools is that they need to be in the hands of the right people," he said.
"We want to make sure that the forces that are establishing order in our democracies have these tools, and the forces that want to disrupt us or commit crime, do not."
The UK banks which will now gain access to OpenAI's GPT 5.5 Cyber include Lloyds Banking Group, HSBC and Nationwide. NatWest and Santander already have access under existing agreements.
According to a company representative, Anthropic is urgently working to expand access to Mythos. They believe that Mythos' capabilities sit in a tier above 5.5 Cyber and therefore more caution is required.
Anthropic initially opened it up to a collective of 42 companies, mostly other US tech firms.
OpenAI and UK sign deal to use AI in public services
OpenAI has been more generous with access to Cyber 5.5, opening it up in the EU and also to banks in Japan and Canada, among others.
Both firms charge for usage although Anthropic put $100m towards Mythos previews.
"I don't necessarily expect these tools to surface things humans wouldn't find eventually but they are relentless and incredibly thorough in sorting through the millions of lines of code which are in banking apps alone," said Prof Alan Woodward, a cybersecurity expert from Surrey University.
"In the UK, some of the code which banking systems still run on is incredibly old, and AI will be very useful at finding problems there.
"They can do many weeks' worth of work in minutes – but they do also surface false positives, so you still need a human in the loop to check their findings."
What is AI and how does it work?
Finance ministers and bankers raise serious concerns about Mythos AI model
Anthropic investigating claim of unauthorised access to Mythos AI tool
