This episode looks at Anthropic's new Mythos model, which has already uncovered thousands of previously unknown vulnerabilities across major operating systems and browsers. The model's speed and accuracy show how quickly offensive grade AI is evolving and what that means for firms that still rely on traditional security reviews.

This matters for leaders in law, accounting, consulting, and other professional services because the attack surface is expanding faster than most firms' ability to monitor it. If AI can find flaws in code from the largest tech companies, then every firm needs to assume its own systems contain gaps that older tools will miss.

We also cover Russian router hijacks targeting Microsoft three sixty five, new court rulings on AI use in legal work, and why many billing disputes point to structural revenue issues, not client behavior.

Learn more at https://crestvale.io

Support the show