Anthropic Hits the Brakes, ShinyHunters Breaks Trust and China Builds AI
AI is pushing deeper into safety, security and national infrastructure all at once.
On this episode of Techstrong Gang, hosts Alan Shimel and Mike Vizard, joined by Fred Wilmot, Teri Robinson and Jon Swartz, break down Anthropic’s new hard safety brakes for Mythos AI, the trust-based tactics behind ShinyHunters’ success and China’s reported $295 billion AI investment focused on an interconnected network.
The first segment, AI Safety Brakes, looks at Anthropic’s effort to put stronger safeguards around a more powerful AI model. But if Mythos can still generate N-day exploits in minutes or hours, then the question becomes whether safety brakes can truly keep up with capability.
The second segment, ShinyHunters Success Story, turns to one of the most important lessons in cybersecurity: attacks often succeed not because defenses are absent, but because trust gets weaponized. The panel explores how ShinyHunters keeps finding leverage by crossing that trust barrier.
The final segment, China AI Utility, examines what it means when AI becomes part of a nation-scale utility strategy. China’s reported investment signals a future in which AI infrastructure is treated less like a market experiment and more like foundational public capability.
Taken together, these stories point to a broader shift: AI is no longer just a software story. It is a safety story, a trust story and an infrastructure story.
