AI Model Update Boosts Automated Vulnerability Scanning for Businesses

Claude has announced that its newest AI model, Mythos 5, is now integrated into Claude Security, allowing organisations on Claude Enterprise plans to scan their codebases for vulnerabilities and receive suggested patches. The company says it is also working with cybersecurity technology and service partners to bring the model into other tools that defenders already use, aiming to make advanced AI-driven threat detection more widely accessible.

Alongside the model rollout, Claude launched a $35 million Defender Advantage Fund, offering credits to organisations that work on patching vulnerabilities in open-source software and automating scanning and patching processes. The company noted that it is designing access carefully: rather than giving users direct, unrestricted access to the AI model, it is focusing on delivering specific, safer outputs such as vulnerability patches or security alerts, which reduces the risk of the technology being misused by malicious actors.

For small and medium businesses, these kinds of AI-assisted security tools could eventually make vulnerability detection and patching more affordable and accessible, particularly for organisations that rely on open-source software components but lack dedicated security teams.