Visa aimed Anthropic's Claude Mythos on the infrastructure behind billions of each day transactions, a community that spans greater than 200 international locations and territories, strikes cash in roughly 160 currencies, and connects almost 5 billion fee credentials to greater than 175 million service provider places.
The mannequin stitched minor weaknesses deep within the stack into working exploit chains that might historically have surfaced solely late in penetration testing. Rajat Taneja, Visa's president of expertise, walked the VB Rework 2026 viewers by what got here subsequent, together with why Visa launched the harness that ruled your complete hunt as open supply and why the corporate deserted conventional remediation metrics for a measurement its group invented.
Taneja has run expertise technique, product engineering, and international infrastructure at Visa since 2019, after becoming a member of the corporate in 2013 from Digital Arts, the place he served as CTO following 15 years at Microsoft. He co-authored, with Visa chief data safety officer Subra Kumaraswamy, the June 10 weblog put up asserting the discharge of the Visa Vulnerability Agentic Harness on GitHub as a reference implementation that any safety group can examine, adapt, and prolong. Visa additionally printed a technical white paper detailing the structure, classes realized, and 12 non-negotiable architectural practices for crucial infrastructure.
Belief constructed on pessimism and paranoia
Taneja led with the arithmetic that makes Visa a goal value defending obsessively. Belief on the scale of world funds will get engineered by what he known as pessimism and paranoia, by assuming failure and designing round it earlier than failure arrives. The community has been hardened over a few years by zero-trust structure, layered defenses, and extremely automated safety operations constructed for the size and reliability international funds demand.
So when Anthropic invited the organizations behind crucial software program to check Mythos underneath Mission Glasswing, Visa stated sure. Glasswing members collectively recognized greater than 10,000 high- or critical-severity vulnerabilities within the first month of testing throughout software program underpinning crucial programs industry-wide, in response to Anthropic. Anthropic's personal conclusion positioned the bottleneck after discovery, in verification, disclosure, and patching velocity. Visa joined to check a long time of hardening at AI velocity and study the place superior fashions might push its defenses additional.
What Mythos confirmed at Visa
Inside Visa's surroundings, Mythos demonstrated system-wide, context-aware evaluation, surfacing vulnerabilities buried deep within the stack and flagging points that develop extra critical when chained collectively, with findings clear sufficient that engineering groups might act on them with out wading by noise. Some findings carried crucial severity scores, and Visa credit its zero-trust controls, community segmentation, and layered safeguards with breaking the chain earlier than any exterior actor might have acted.
That affirmation mattered, Taneja stated, however the epiphany that adopted mattered extra. "In a world of agentic assaults, protection additionally must be agentic," he stated. Even at an organization that has invested a long time in defense-in-depth, the mannequin revealed assumptions the group had been working underneath that wanted rethinking. Conventional SAST instruments preserve their place as a primary move towards recognized vulnerability patterns, Visa's white paper notes, however sample matching alone can’t observe an adversary who causes by logic, knowledge movement, and the exploit chains that dwell between the signatures.
A harness, not a scanner
Visa's response was not one other monolithic scanner. The group constructed the Visa Vulnerability Agentic Harness, now in its fifth era, as a ruled pipeline that directs frontier AI fashions by structured safety duties whereas implementing deterministic controls, coverage gates, and human oversight at each stage. Taneja walked by the design philosophy. The harness operates throughout 4 phases and eleven levels, from code ingestion and menace modeling by deep-dive verification, exploit chain synthesis, and at last remediation and repair validation.
Three design selections drive discovering high quality, per the venture's personal documentation. Menace modeling runs earlier than evaluation to deal with the assault floor reasonably than scanning all the things blindly, multi-agent deterministic voting requires convergence throughout impartial reasoning chains earlier than a discovering advances, and structured triage artifacts compress the lifecycle from discovery to a consequence builders can truly ship. The payoff is a pipeline that runs sizzling by default. A plain scan within the shipped profile runs all eleven levels and edits supply information within the goal repository in repair mode, making use of candidate patches except the operator stops it at detection.
The harness is multi-model by design. An LLM abstraction layer lets Visa swap or mix suppliers with out altering the management airplane, and the open-source model works with Anthropic Claude, OpenAI-compatible fashions, or a mixture. The repo's documentation is candid in regards to the exception. Making use of a repair requires the file-editing instruments that solely the Anthropic backends expose, so the remediation and validation levels presently require Anthropic fashions for full performance, and an OpenAI-compatible mannequin in these roles is proscribed to report-only output. VentureBeat's Q2 2026 Pulse analysis, offered earlier on the convention, reinforces why that supplier flexibility issues. Among the many enterprises surveyed, 82% depend on provider-native controls as their main safety layer, and 59% plan to undertake or swap agent safety tooling inside the yr. The controls enterprises adopted final yr are already changing into the controls they plan to switch.
Imply Time to Adapt replaces legacy metrics
Discovering vulnerabilities is now not the onerous half, Taneja argued. The actual problem is how shortly a group can affirm a difficulty is really exploitable, repair it, and show the assault path is closed reasonably than simply displaying a patch was utilized. Visa calls this Imply Time to Adapt, and the white paper tracks it alongside three dimensions. Stock freshness measures how present and full the group's view is of code, configuration, and runtime deployment. Exploitable paths per launch counts what number of end-to-end assault chains stay potential after every launch, not simply what number of findings had been closed. Validation cycle time tracks how lengthy it takes to supply repeatable, evidence-backed proof {that a} repair works and stays working in manufacturing.
That distinction issues as a result of legacy measures similar to imply time to detect and uncooked CVE closure counts can look higher on paper whereas precise publicity retains rising beneath them. A company can shut a whole lot of findings a month and nonetheless depart viable exploit chains open if no one examined whether or not the patches truly break the assault. MTTA forces groups to measure the result that issues, and the white paper leans on CISA Recognized Exploited Vulnerabilities knowledge to make the prioritization case, noting that fewer than 1% of CVEs are ever actively exploited. Visa's SSDLC coverage now assumes each exploitable path can be exercised in manufacturing and requires it to be remediated earlier than code is promoted.
Provide chain danger accelerates underneath AI
The dialog moved previous Visa's personal perimeter when Taneja turned to suppliers. A well-defended enterprise stays uncovered by weak distributors and weak open-source parts, the white paper warns, so Visa is making AI-specific safety posture a non-negotiable dimension of provider due diligence, with expectations for steady vulnerability validation, dwelling software program payments of supplies, and MTTA baselines throughout its expertise stack.
Visa has additionally joined Mission Lightwell, the $5 billion IBM and Purple Hat initiative to harden broadly used open-source parts by AI-driven validation and coordinated patching, alongside monetary establishments together with Financial institution of America, JPMorganChase, Goldman Sachs, and Mastercard. The dedication extends the identical logic upstream, as a result of the MTTA clock doesn’t pause at any single firm's perimeter.
When brokers begin shopping for issues
Securing agentic commerce is Visa's subsequent downside. Taneja described a future the place AI brokers transact on behalf of customers and enterprises, and stated Visa is constructing the belief framework, identification layer, and agent readiness scoring that retailers will want earlier than brokers can safely full transactions. Behind that work sits the Visa Fee Threats Lab, a simulation surroundings the place actual fraud situations get replayed towards the authorization guidelines, thresholds, and configurations Visa truly runs, to floor AI-enabled failure modes as focused hardening suggestions.
The identification problem just isn’t theoretical. VentureBeat's Pulse analysis discovered that 69% of enterprises already run credential sharing someplace of their agent deployments, and firms with shared credentials report safety incidents or near-misses at a 63.5% price, towards 40.9% the place each agent has its personal scoped identification. Visa's white paper addresses that hole straight, itemizing "AI brokers are identities" amongst its 12 non-negotiable practices and requiring scoped permissions, least privilege enforcement, full audit trails, and inclusion in IAM governance for each agent that calls an API, reads knowledge, or modifies a system.
Three priorities for defenders
Visa is organizing its defensive technique round three priorities, Taneja stated. Shift safety left till exploitable flaws are designed out earlier than they attain manufacturing, and change high-risk, under-supported parts earlier than they flip into materials publicity. The third is the heaviest carry at Visa's scale, refactoring defenses to run autonomously underneath human governance so detection, validation, and response preserve tempo as menace quantity grows and the fashions behind assaults enhance.
None of it requires a fee community's funds to start out. The harness sits on GitHub with 595 stars and 97 forks as of July 20, MTTA wants a dashboard reasonably than a procurement cycle, and the white paper's 12 non-negotiable practices map onto structure opinions safety groups already run. Visa's personal conclusion reads like a deadline. The opening to get forward of machine-speed attackers remains to be there, the paper argues, and it’ll not keep open.
