In addition to model usage credits, Anthropic donated $2.5 million to Alpha-Omega and OpenSSF through the Linux Foundation, and $1.5 million to the Apache Software Foundation “to enable the maintainers of open-source software to respond to this changing landscape.”
“Partners will, to the extent they’re able, share information and best practices with each other; within 90 days, Anthropic will report publicly on what we’ve learned, as well as the vulnerabilities fixed and improvements made that can be disclosed,” Anthropic stated.
Anthropic said it will collaborate with security organizations to produce a set of practical recommendations AI security practices such as vulnerability disclosure, software update, open-source and supply-chain security, and software patching automation.
“Claude Code is changing how developers build software. AI agents are reshaping how enterprises automate operations,” wrote CrowdStrike in its own blog post about the project. “Anthropic’s Mythos Preview expands the reasoning, planning, and execution capabilities of AI agents. They all touch the endpoint — where data is accessed, decisions are made, value is delivered, and risk is born.”
“New models are also where opportunity is the largest. The same frontier models that expand the attack surface give defenders a capability advantage that did not exist a year ago: discovering vulnerabilities, detecting threats, and responding to incidents faster than ever before,” CrowdStrike wrote.
CrowdStrike’s 2026 Global Threat Report found an 89% increase in attacks by adversaries using AI year-over-year. The use of AI for vulnerability discovery and exploit development is accelerating on both sides, the vendor stated.
Click Here For The Original Source.
