Anthropic’s Project Glasswing: A New Initiative to Secure Critical Software

On April 7th, 2026, AI company Anthropic announced Project Glasswing, which is an effort to protect critical software. It’s powered by Claude Mythos Preview, Anthropic’s most capable model yet. Mythos Preview has identified thousands of high-severity vulnerabilities, and considering the rate of AI growth, the project is focusing on finding ways to protect software from these risks as AI expands.

What is Project Glasswing?

Project Glasswing brings Amazon Web Services, Anthropic, Apple, Broadcom, Cisco, CrowdStrike, Google, JPMorganChase, the Linux Foundation, Microsoft, NVIDIA, and Palo Alto Networks together to secure the most critical software in the world. These partners will use Mythos Preview for defensive security work, and Anthropic will share information from this project so the entire industry can benefit.

Anthropic is committing up to $100 million in usage credits for this project, along with $4 million in direct donations for security organizations. Project Glasswing is only a starting point for addressing security risks. Cybersecurity problems can’t be solved by just one organization, so Anthropic’s efforts will continue to grow and change as they gain new information throughout this initiative.

Addressing Cybersecurity Vulnerability

Cybersecurity risks exist in most of the everyday software we use, including banking systems, medical information storage, corporate networks, and energy infrastructure. AI has made impressive advances to spot vulnerabilities and find ways to handle them, which is why Anthropic is using Claude Mythos Preview to identify thousands of vulnerabilities.

Here are a few of Mythos Preview’s discoveries:

• A 27-year-old vulnerability in OpenBSD, which is known for being one of the most security-hardened operating systems worldwide.
• A 16-year-old vulnerability in FFmpeg, which is used by countless pieces of software for encoding and decoding video.
• Several vulnerabilities in the Linux kernel, which runs most of the world’s servers.

If vulnerabilities can exist for so long in major software, there are likely lots of cybersecurity concerns that are currently going unnoticed. Project Glasswing’s goal is to locate these vulnerabilities so we can find the best way to protect people from them.

Future Plans for Project Glasswing

The announcement of Project Glasswing is only the beginning of long-term efforts. Currently, partners of this project will use Claude Mythos Preview to find and fix vulnerabilities and weaknesses identified in their foundational systems. While this is a great starting point, the project will eventually need broad involvement across the technology industry to be fully successful.

Anthropic plans to grow this project over time, sharing as much information as possible throughout. After the first 90 days, Anthropic will release a public report about everything they’ve learned so far to ensure everyone has the information they need to handle risks. Anthropic has also been discussing this project with U.S. government officials to strengthen the impact.

Keep Your Brand Protected

Cybersecurity concerns pose huge risks for all types of brands. Keeping an eye on this new information can help brands protect their software.

While focusing on cybersecurity efforts, don’t forget to stay on top of marketing tactics to keep your brand visible. Contact Avenue Z today to experience how we can elevate your revenue and reputation.