Anthropic launched the Anthropic Cyber Mission, a long-term initiative to help secure critical infrastructure and open-source software by providing defenders with AI (artificial intelligence) tools, research and technical resources. The effort includes the Critical Infrastructure Defense Program (CIDP), which combines Anthropic’s frontier Claude models, on-site engineers and threat research to help protect OT (operational technology) supporting power grids, water systems, factories and transportation networks.
The program’s founding partners include Accenture, Booz Allen, CrowdStrike, Deloitte, Dragos, Hitachi, Insane Cyber, Nozomi Networks, Palo Alto Networks, PwC and Rockwell Automation. Anthropic said several partners are already working with Claude to address vulnerabilities and help customers fix them, beginning with a small group of providers to assess which approaches are most effective and practical.
“These organizations represent the providers that operators rely on to secure these systems: the consulting and advanced technology companies that run security programs, the security companies that guard the business and industrial networks, and the manufacturers that build and patch the equipment itself,” according to an Anthropic announcement last week. “This work is underway. Several partners are currently working with Claude to fix vulnerabilities and help customers do the same.”
It added that critical infrastructure is hard to defend in many ways that AI cannot fix, but we believe that frontier models can help find and repair weaknesses before those weaknesses are used to cut off power or make water unsafe. Our first step is to work with a small cohort of providers to learn which strategies are most effective and practical.
The initiative addresses the security challenges posed by legacy OT systems, which often cannot be taken offline for patching and may retain known vulnerabilities for years because of operational constraints, proprietary equipment and safety risks. Anthropic said state-sponsored adversaries have spent years gaining footholds in critical systems, while the increasing availability of highly capable AI models has reduced the cost of exploiting vulnerabilities.
The San Francisco, California-based company also highlighted difficulty of verifying, prioritizing and fixing security findings, noting that vulnerabilities identified during Project Glasswing sometimes remained unresolved for months, while fixes to OT systems can, in rare cases, take decades to implement safely. Anthropic said the Cyber Mission will expand to more partners and sectors, to reduce exploitable entry points and enable faster recovery from attacks on essential services.
“As part of Project Glasswing, partners uncovered many vulnerabilities, but we haven’t yet achieved a sufficient reduction in cyber risk. It’s easier than ever to find vulnerabilities, but verifying, prioritizing, and fixing these findings remains challenging,” the post added. “Based on lessons from Project Glasswing, we’re starting these new efforts to support defenders. Earlier this week, we merged Project Glasswing into our expanded Cyber Verification Program, which gives many more defenders access to our most capable models. And through the Anthropic Cyber Mission, we’ll deploy Anthropic engineering talent and provide tools and funding for those who secure critical infrastructure and open-source software.”
In June, Anthropic launched a cyber defense program for state, local, tribal, and territorial governments. “Since then, we’ve offered frontier Claude models and technical support to more than half of all US states and some of the country’s largest public critical infrastructure operators, speeding up code scanning and patching, incident response, red teaming, and other security workflows. Our approach is to work with the companies, coalitions, governments, and nonprofits that know this landscape and that operators trust, and provide support where it’s most needed.”
Anthropic said almost all software relies on open-source code, much of it maintained by small teams of volunteers. Under Project Glasswing, the company scanned hundreds of widely used open-source projects, had humans triage and review many of the potential vulnerabilities uncovered, and privately reported its findings to maintainers through its coordinated vulnerability disclosure process. Some maintainers with the capacity to triage at scale asked for everything the company’s models had found in their projects, reviewed or not. In response, Anthropic is launching OSS Scanner, an opt-in service inspired by Google’s OSS-Fuzz, which gives enrolled projects periodic scans from its most capable models free of charge.
Each report includes a proof of concept showing how the bug could be exploited, an explanation, and a suggested fix where one is available. The reports are model-generated and sent without human review, which means maintainers receive them faster but also that some will contain inaccuracies, such as a wrong severity rating. Anthropic said it expects a true-positive rate above 90% and will work to improve that rate and the quality of the fixes over time. The service is meant for projects with the capacity to keep up with surfaced findings, and for other projects the company will continue to share human-verified disclosures under its coordinated vulnerability disclosure policy.
Anthropic described OSS Scanner as a first step and said its open-source work will aim to get findings to maintainers faster, accelerate fixes and explore new secure architectures and coding practices. To get findings out faster, it plans to bring OSS Scanner to more projects while keeping human-verified disclosure for those that need it. To accelerate fixes, it plans to automate triage and patching, which are still mostly manual, for projects that want it. It also plans to research and share methods for hardening or rewriting code with projects that want to go further than patching. The company said it will proceed based on guidance from open-source maintainers and foundations and will share what works so other projects can replicate it.
Anthropic said it has funded the organizations behind widely used open-source code, including the Python Software Foundation, Alpha-Omega and the OpenSSF through the Linux Foundation, and the Apache Software Foundation. It is also supporting Akrites and Gold Eagle, which collect and coordinate vulnerability reports from many sources to avoid overwhelming maintainers. The Defender Advantage Fund (0xDAF), which the company launched in August, supports pilot programs in these areas and keeps OSS Scanner free. Maintainers can also apply to Claude for Open Source for free Claude Max subscriptions, and to the Cyber Verification Program for expanded access to Claude’s cyber capabilities for defensive work.
Anthropic said highly cyber-capable AI models are now widely available to attackers, while defensive tools, including its own, have not yet reached enough of the defenders who need them. The company pointed to OSS Scanner, its Cyber Verification Program, defensive products that partners build on its platform, and Claude Security as some of its efforts to close that gap, and said it will add others as soon as possible.
The company forecast that in two years AI will favor defense, making it easier to catch bugs before they ship, write fundamentally secure software from scratch and actively defend systems with models. In the near term, however, that may not hold true. The cost of exploiting vulnerabilities has dropped, while verifying, disclosing and fixing them remains slow and still depends on people. Anthropic said it often saw months pass between a vulnerability being found and being fixed during Project Glasswing. For operational technology, a fix may have to wait until it can be applied safely to running machinery, which in some rare cases might take decades.
Anthropic said success means critical infrastructure such as water, power, transport and communications keeps running in the face of attacks by capable adversaries, with fewer exploitable paths in and faster recovery when something gets through. The company said this will take years and that it expects the Anthropic Cyber Mission to change as it learns, together with the public and private sectors, what works.
“Over the coming months, we’ll bring the Critical Infrastructure Defense Program to more partners and more sectors, and we’ll share what we learn along the way, including what didn’t work,” the Anthropic announcement mentioned. “We also expect other AI developers, security companies, and governments to run efforts of their own, and we’d like to collaborate with them wherever that helps defenders.”
It added, “We will also expand our work to help secure open source and the broader supply chain. This includes researching new ways of writing software and new ways of defending systems. We intend to share frontier research, tools, and resources to help others secure their software and supply chain.”
Last November, Anthropic warned that cybersecurity has reached a critical inflection point, with AI models becoming powerful tools for both defensive and offensive operations. The warning comes after China’s state-sponsored hackers reportedly used Anthropic’s AI technology to automate intrusions into major corporations and foreign governments during a September hacking campaign.
