Visa Expanded Cybersecurity Services and Agentic Framework

The payments giant updated its model-agnostic security framework to automate vulnerability remediation and validation.

Updated on Sept. 22, 2026 in Cybersecurity

Isometric editorial illustration of a complex interlocking steel lattice structure, symbolizing automated cybersecurity remediation systems. — Visa updated its open-source vulnerability harness to include automated remediation, aiming to reduce the response time between identifying and patching cybersecurity threats. AI Illustration. Upload story photo >

Live Poll

Do you trust that major companies are effectively using AI to protect your personal information?

Visa has updated its open-source Visa Vulnerability Agentic Harness framework to include automated remediation and validation capabilities. The company also introduced three new cybersecurity advisory services to help organizations address risks accelerated by artificial intelligence.

Why it matters

As artificial intelligence accelerates the discovery of vulnerabilities, organizations face an urgent need to shorten the response time between identifying a threat and implementing a fix. These tools aim to stabilize security workflows by minimizing the gap before exploitation occurs.

The updated framework supports both Anthropic and OpenAI models and enables closed-loop remediation where structured feedback automatically refines failed security fixes. Organizations can now deploy these AI models through configuration updates rather than custom code changes.

The players

Visa

A global payments technology company that facilitates electronic funds transfers and operates complex financial transaction networks.

NVIDIA

A designer of graphics processing units and specialized AI hardware that leads the industry in infrastructure for large-scale machine learning models.

IBM

A multinational technology corporation specializing in enterprise cloud computing, hybrid software, and AI-driven business solutions.

Red Hat

A provider of open-source software solutions, including the Red Hat Enterprise Linux platform, currently a subsidiary of IBM.

The details

The Visa Vulnerability Agentic Harness is a software framework that automates the identification and patching of security flaws. It features a closed-loop system, a process where output is fed back into the input to improve results, which allows the AI to refine failed attempts at remediation without requiring a full restart of the security cycle. By supporting model-agnostic architectures, the system allows teams to swap between AI providers based on their specific security requirements.

June 2026: Visa released the original Vulnerability Agentic Harness framework as open-source.
September 22, 2026: Visa announced the portfolio enhancements and new advisory services.

The Tech Race

Visa is integrating its security tooling with broader industry efforts, such as the NVIDIA Open Secure AI Alliance and the IBM and Red Hat Project Lightwell. These moves place the company at the center of a collaborative push to establish standardized safety protocols for deploying AI models in high-stakes financial environments.

Organizations utilizing the Visa framework can now shift their cybersecurity workflow toward model-based configuration, reducing the need for manual code-heavy intervention. The new advisory services are available immediately for enterprises seeking to modernize their security posture against AI-driven threats.

The takeaway

Visa is shifting from reactive patching to an automated, agentic model of security that prioritizes speed and feedback loops. Industry observers should watch for the next wave of model integrations within the open-source harness as it continues to scale beyond its current user base.