Visa opens part of AI cyber defence system to the public
In short
Visa has made part of its AI cyber defence system open source after problems exposed by AI models. Rajat Taneja said the move followed lessons from Anthropic's Mythos AI model earlier this year.
It matters because Visa is sharing defensive tools more widely, but the report does not say how big the security gaps were or what risks remain.
What the reporting says
Visa open-sourced part of its AI-powered cyber defence system after weaknesses exposed by AI models highlighted vulnerabilities, according to a Reuters-linked summary. Visa President of Technology Rajat Taneja said the company made the system available after lessons from Anthropic's Mythos AI model earlier this year. The immediate consequence is a wider public release of some defensive tools, while the full scope of the exposure is not stated in the available reporting.
Why it matters
Visa open-sourced part of its AI-powered cyber defence system after weaknesses exposed by AI models showed vulnerabilities, according to a summary citing President of Technology Rajat Taneja. The move suggests Visa is adapting its security approach after the issue surfaced earlier this year through Anthropic's Mythos AI model. The report does not specify which parts were released, what was fixed, or whether other systems were affected.