The Artificial Intelligence Show #209: Claude Mythos, Project Glasswing, Claude Code Leak, OpenAI Raises $122B & the End of Middle Management
578 snips
Apr 14, 2026 A mysterious Anthropic model sparks cybersecurity alarms and national security fears. A code leak and shifting subscriptions raise questions about control as AI races ahead. OpenAI’s giant funding round collides with scrutiny, politics, and real-world attacks. The conversation also explores job disruption, regulation, supply chain breaches, and AI becoming a personal knowledge layer.
AI Snips
Chapters
Transcript
Episode notes
The Mercor Breach Exposed Hidden AI Supply Chain Risk
- The Mercor breach shows how AI supply chains create hidden systemic risk well beyond the major labs themselves.
- Attackers allegedly pulled four terabytes including contractor data, source code, and possibly evidence of how labs like OpenAI and Meta train frontier models.
Karpathy Turned LLMs Into A Second Brain
- Andrej Karpathy's markdown wiki workflow hints that LLMs may become persistent knowledge infrastructure, not just chatbots or coders.
- He dumps raw documents into folders, has an LLM compile and link them into Obsidian, then periodically checks for inconsistencies and updates.
The AI Jobs Debate Is Turning Mainstream
- AI job anxiety is moving from fringe concern toward mainstream acceptance among economists, employers, and students.
- Challenger counted about 15,000 March job cuts tied to AI, Jack Dorsey argued for replacing middle management, and 42% of bachelor's students reconsidered majors.