OpenAI launched a Trusted Access program giving Microsoft its most capable models to bolster cybersecurity. This move counters the rise of autonomous flaw-hunting models like Anthropic's Mythos. The partnership focuses on defending against AI-driven threats. Security practitioners can now leverage higher-tier model capabilities to identify and patch vulnerabilities faster than previous iterations allowed.