Policy dependency / Stack layer
David Sacks Publishes a Six-Point Rebuttal to Dario Amodei's Trust Essay, Renaming His 'FINRA for AI' Proposal the 'DMV for AI'
X/Twitter
Stack layer / Contrast
Qwen3.8 27B Scores 52 on Artificial Analysis' Intelligence Index — #1 of 135 Open-Weights Models, Apache 2.0, 256K Context
Artificial Analysis / Hacker News
Stack layer / Contrast
'How Do Agents Fail on AutoResearch': 800 Trajectories Across 8 Harness-Model Combos Find One Failure Every Model Shares — They Never Check Output Against Evidence
arXiv (via HuggingFace Daily Papers)
Policy dependency / Stack layer
In the Same Post, Amodei Rejects the Regulation-vs-Open-Models Binary and Endorses a FINRA-Style Self-Regulator for AI
TechCrunch
Policy dependency / Threat pattern
MIT Tech Review: the Flock debate is about 120,000 cameras' design choices, not whether they catch criminals
MIT Technology Review
Contrast / Update thread
Proof-of-Execution Memory: one LLM rewrite defeats the SENTINEL memory defense entirely, and stronger models are more vulnerable
arXiv
Stack layer / Update thread
CROSS-CATEGORY: MCP Is Quietly Replacing the Vendor Dashboard — ElevenLabs, ZoomInfo and S&P Global All Shipped Connectors That Make Their Own UI Optional
ElevenLabs (corroborated by The New Stack, ZoomInfo IR, and S&P Global/Microsoft announcements Aug 11-12)
Stack layer / Threat pattern
Greg Brockman calls the OpenAI–Hugging Face agent breach a "watershed moment" and publishes a 10-step defender playbook
Greg Brockman / OpenAI