Episode 15 - The Right Eyes: Mythos, and the Future of Vulnerability Discovery

The emergence of advanced large language models like Anthropic's Mythos represents an epochal shift in cybersecurity, fundamentally altering how zero-day vulnerabilities are surfaced and remediated. In this episode, host Richard Bejtlich sits down with Corelight Co-founder Greg Bell to analyze the security implications of this AI-driven bug explosion, highlighting recent AI-assisted vulnerability discoveries across infrastructure mainstays like FreeBSD and Firefox. Together, they challenge the classic open-source maxim that "with enough eyes, all bugs are shallow," arguing instead that the arrival of the right automated eyes exposes an overwhelming pool of latent software flaws. Moving beyond the immediate operational chaos, Richard and Greg discuss the economics of declining token costs, the critical survival need for an assume-breach mentality, and how Corelight’s new agentic triage capabilities help defenders automate mind-numbing log review to achieve a resilient, human-led cybersecurity equilibrium.