Rockwell Automation is testing advanced artificial intelligence to find and fix security weaknesses in the software and connected systems that underpin factories and other critical infrastructure.
The industrial automation company has joined Anthropic’s Project Glasswing, an initiative that gives approved organizations controlled access to powerful Claude models for defensive cybersecurity research. Rockwell’s security teams will use Claude Mythos 5 to explore faster methods for discovering, validating, prioritizing, and remediating vulnerabilities.
The collaboration targets a growing challenge for industrial operators. Manufacturers are connecting production equipment, operational technology, cloud platforms, and business systems at an accelerating pace. That convergence can improve efficiency, but it also creates more pathways for attackers to disrupt physical operations.
AI could help defenders analyze large codebases, identify potential vulnerabilities, test whether suspected flaws are exploitable, and determine which problems demand immediate attention. For industrial organizations, faster prioritization is especially important because patching production systems can require scheduled downtime, extensive testing, and coordination across complex supply chains.
“Rockwell Automation plays a central role in helping manufacturers operate safely, securely, and reliably,” said Tony Baker, vice president and Chief Product Security Officer, Digital Trust, Rockwell Automation. “By participating in Project Glasswing, we are applying advanced AI capabilities in a controlled, defensive way to help identify and address vulnerabilities faster and support the resilience of the industrial ecosystem.”
Anthropic launched Project Glasswing in April 2026 with approximately 50 partners. The program has since expanded to include about 150 additional organizations across more than 15 countries, according to the company. Participants include organizations supporting power, water, healthcare, communications, hardware, and other critical sectors.
The project is built around restricted access and defensive use cases, reflecting the security industry’s broader effort to capture the benefits of highly capable AI models without providing unrestricted access that could facilitate offensive operations.
For Rockwell, the partnership could shorten the time between vulnerability discovery and remediation while generating security insights applicable across its industrial product portfolio. The larger test will be whether AI-assisted vulnerability research can deliver meaningful improvements without overwhelming security teams with uncertain findings.
As AI becomes more capable of analyzing software and identifying attack paths, industrial cybersecurity may become one of its most consequential proving grounds. In environments where digital compromises can halt production or affect physical infrastructure, faster defense carries unusually high stakes.
