CrowdStrike has joined Anthropic’s Critical Infrastructure Defense Program to help strengthen AI-powered cybersecurity across energy, water, manufacturing, and transportation systems.
Announced on October 8, 2026, the partnership combines CrowdStrike’s experience protecting business and industrial networks with Anthropic’s advanced Claude models, technical support, and cybersecurity research.
The initiative targets a difficult security challenge. Critical infrastructure often connects modern information technology, or IT, with older operational technology, or OT, that controls physical equipment.
These industrial systems can remain in service for decades. Taking them offline for updates may interrupt essential services, leaving known security weaknesses unresolved for years.
Anthropic Defense Program
According to Anthropic’s announcement, the program brings Claude models, on-site engineers, and threat research to trusted security providers and equipment manufacturers.
CrowdStrike is among 11 founding partners, alongside organizations including Dragos, Nozomi Networks, Rockwell Automation, Palo Alto Networks, Accenture, and Deloitte. The approach puts AI support in the hands of teams that already understand industrial environments.
The program forms part of the broader Anthropic Cyber Mission, which also supports open-source software security. Anthropic says several partners are already using Claude to fix vulnerabilities and help customers address weaknesses.
Its initial focus is learning which defensive methods work safely and effectively, rather than treating AI as a complete solution for industrial security.
CrowdStrike will contribute security data, threat intelligence, and experience from its Falcon platform. In its partnership announcement, the company describes an opportunity to connect its knowledge of attacker behavior with Anthropic’s AI research.
That context matters because defenders must understand a weakness within the wider network before they can choose a practical response.
A key part of CrowdStrike’s contribution is Falcon for XIoT, which covers industrial and connected devices alongside IT assets. CrowdStrike says it helps discover unmanaged and older equipment, assess exposure without disruptive scanning, and detect threats moving between business and operational networks.
Supported Windows and Linux hosts can help discover and monitor nearby unmanaged devices. The Falcon platform also combines endpoint, identity, cloud, threat intelligence, and OT security data. This shared view gives defenders more context when investigating suspicious activity across connected environments.
The partnership builds on an existing relationship, with Anthropic’s Claude already available within CrowdStrike’s Charlotte AI security assistant. Participation in the program adds technical guidance and access to a community focused on critical infrastructure defense.
The announcement also follows Anthropic’s expanded Cyber Verification Program, which gives eligible security professionals broader model access for approved work. However, finding weaknesses is only part of the challenge.
Anthropic acknowledges that verifying findings and safely applying fixes remain difficult, especially when industrial equipment cannot stop. The collaboration’s value will depend on turning AI findings into safe, practical improvements.
An EDR alert reveals an attack. But can it tell the full story? Discover how to reconstruct the attack timeline.