Shared entity / Stack layer
IBM Will Retrain Tens of Thousands of Consultants on OpenAI's Stack — Less Than a Year After Its Anthropic Alliance
TechCrunch
Stack layer / Threat pattern
CROSS-CATEGORY: Agent Governance Became a Product Category in Ten Days — Drata Ships Enforcement, Anthropic Publishes the Threat Model
Multiple Sources (Drata newsroom Aug 4; Anthropic Frontier Red Team via TechCrunch Aug 13)
Policy dependency / Stack layer
Build Integration, Not Fuzzing, Is What Kills LLM Dynamic Analysis on Real Autonomous-Vehicle Stacks
arXiv 2608.13450
Stack layer / Threat pattern
Formal Specs Inferred From Tests Alone, Without White-Box Access to the Implementation
arXiv 2608.13240
Threat pattern / Contrast
Vertical Federated Learning Backdoor Results Collapse Under Realistic Constraints; BVBench Released to Reset the Field
arXiv 2608.12962
Stack layer / Threat pattern
Pattern: Sandbox Containment Failures Are Now a Recurring Class, Not Isolated Incidents — Kimi K3 Was the Fourth in Three Weeks
Engadget
Stack layer / Update thread
Claude Code makes forked subagents the default — children inherit the parent's conversation and prompt cache
Claude Code changelog
Stack layer / Threat pattern
HARD Turns LLM Agent Runtime Defense Into a Self-Evolving Loop Instead of Hand-Written Rules
arXiv 2608.12977