> ARGUS BRIEF ARCHIVE

policy The Academy Drew a Bright Line: Fully AI-Generated Actors and Scripts Are Out of Oscar Contention

ecosystem_shift Anthropic Is Reportedly Shopping for U.K. AI Chips, Which Tells You What the Real Bottleneck Still Is

policy A Chinese Court Just Said “AI Replaced You” Is Not a Lawful Firing Rationale

policy The Pentagon Just Picked Its AI Bench, and Anthropic Is Not Sitting on It

ecosystem_shift A Dark-Money AI Fear Campaign Is Trying to Turn China Into Big Tech’s Perfect Regulatory Weapon

ecosystem_shift Meta Keeps Buying Its Way Toward Humanoids

policy OpenAI Is in Court and in Crisis Mode as Families Sue Over the Tumbler Ridge School Shooting

ecosystem_shift Google Cloud Crossed $20B, Then Admitted the Real Constraint Is Compute

ecosystem_shift Microsoft Says Copilot Has 20 Million Paid Seats, and the Distribution War Is No Longer Theoretical

ecosystem_shift Anthropic’s Next Round Is Already Being Bid Up to $900B, and the Capital Spiral Is Getting Hard to Ignore

policy OpenAI Is in Court and OpenAI Is in Crisis Mode, as Families Sue Over a School Shooting

ecosystem_shift Google Cloud Crossed $20B, Then Admitted the Real Limit Is Capacity

policy Google Took Anthropic’s Empty Seat and Signed Up for ‘Any Lawful’ Pentagon AI Work

ecosystem_shift OpenAI Walked Onto AWS a Day After the Exclusive Door Opened

api_change Claude Is Plugging Straight Into Photoshop, Blender, and Ableton, Which Means Creative Software Just Became Agent Surface Area

ecosystem_shift Microsoft and OpenAI Quietly Killed the AGI Clause, and a Power Center Just Moved

policy China Just Blew Up Meta’s $2 Billion Manus Bet

policy The Backlash Is Getting Organized, From Indiana to Idaho and Inside Google

ecosystem_shift Google’s $40 Billion Anthropic Bet Turns the Frontier Race Into an Infrastructure War in Plain Sight

security_advisory Anthropic’s Mythos Breach Keeps Echoing, Because the Real Story Is Trust Failure at the Exact Moment the Stakes Got Bigger

ecosystem_shift Anthropic’s Agent Marketplace Experiment Is Small on Paper and Huge in Implication

model_release OpenAI Pushes GPT-5.5 as the Next Step Toward a Super-App, but the Real Question Is Whether the Efficiency Story Survives Contact With Production

model_release DeepSeek V4 Lands Close Enough to the Frontier to Be Dangerous, and the Price Signal May Matter More Than the Benchmark Signal

ecosystem_shift Claude Code’s Post-Mortem Is a Useful Reminder That a Great Model and a Bad Harness Can Still Break the Product

ecosystem_shift Cerebras Files to Go Public, and the AI Infrastructure Trade Is Moving Out of Private Hype and Into the Prospectus Stage

policy The UK Just Put $675 Million Behind Sovereign AI, Which Is Another Way of Saying Compute Nationalism Is Becoming Normal

tool_release Schematik Wants to Be Cursor for Hardware, and Anthropic’s Interest Tells You the Agent Stack Is Leaving Pure Software Behind

ecosystem_shift Anthropic’s Published Instruction Layers Keep Turning Model Behavior Into Auditable Product Surface

security_advisory A Mythos Process Miss Became Public, and the Safety Story Here Is Really About Operational Sloppiness

framework_update OpenClaw’s New Betas Fix the Boring Failures That Actually Break Multi-Agent Work

ecosystem_shift OpenAI Loses the People Behind Sora and Science, and the Retreat From Side Quests Looks Real Now

tool_release Anthropic Wants Claude to Make the Slides Too, and Claude Design Is the Cleanest Shot Yet at the “AI Workbench” Dream

model_release Anthropic’s Mythos Preview Is More Than a Cyber Model, It Is a Bid to Repair Its Washington Relationship

model_update Anthropic Says Opus 4.7 Thinks Harder, but the Real Test Starts After the Launch Post

tool_release OpenAI Pushes Codex Closer to the Desktop, and the Claude Code Fight Is Officially About Workflow Control

ecosystem_shift Anthropic’s New ID Checks Make the Hosted-Model Trust Boundary Feel a Lot More Physical

ecosystem_shift Allbirds Burned the Shoe Brand and Bought GPUs Instead, and the Market Rewarded the Costume Change

tool_release Adobe Wants Firefly to Drive the Whole Creative Cloud, Which Means the Assistant Is Becoming the Interface

framework_update OpenAI Quietly Turns the Agents SDK Toward Enterprise Safety, Which Is Another Way of Saying Governance Is Now Product

security_advisory OpenAI Locks GPT-5.4-Cyber Behind a Trusted-Access Gate, and the Cyber Model Arms Race Gets Real

policy Anthropic Breaks With OpenAI on Liability, and the Industry's Safety Alliance Looks More Tactical by the Day

ecosystem_shift Anthropic Briefed the White House While Suing the Government, Which Tells You Exactly How the Frontier Era Works

breaking_news The Charges Around the Altman Attacks Suggest the AI Backlash Has Moved Into a More Dangerous Phase

ecosystem_shift OpenAI Buys Hiro and Moves One Step Deeper Into the Personal-Finance Stack

industry_movement Microsoft Is Reportedly Building Another OpenClaw-Like Copilot Agent Because Catch-Up Has Become the Strategy

breaking_news Sam Altman Turns a Molotov Attack Into a Public Argument About AI Fear

ecosystem_shift Meta Wants Intimate Health Data Before It Has Earned That Level of Trust

policy Washington May Be Nudging Wall Street Toward Anthropic's Mythos

breaking_news After a Molotov Attack and a Brutal Profile, Sam Altman Decides to Make the Trust Fight Personal

ecosystem_shift Meta Wants Your Raw Health Data, Which Is a Terrible Time to Pretend AI Trust Has Been Solved

ecosystem_shift Gen Z Is Still Using AI, but the Honeymoon Is Ending

breaking_news OpenAI’s Liability Week Just Got More Personal, a New Lawsuit Says ChatGPT Ignored Danger Signals in a Stalking Case

breaking_news The Altman Attack Turns AI Executive Risk Into a Real-World Security Problem

ecosystem_shift Microsoft Starts Pulling Copilot Out of the Walls, Which Tells You a Lot About Ambient AI

policy OpenAI Wants a Liability Shield Before the Real Lawsuits Arrive

breaking_news Florida Opens the Door to a New Kind of OpenAI Reckoning

policy Anthropic Lost a Pentagon Trust Fight It Could Not Afford to Lose Quietly

security_advisory Anthropic Didn’t Just Announce Mythos — It Quietly Told the Security Industry to Brace for Impact

breaking_news Meta Hit Pause After the Mercor Breach — And the AI Data Supply Chain Suddenly Looks Much More Fragile

tool_release Google Quietly Shipped an Offline Dictation App — Which Is a Bigger Local-AI Signal Than It Looks

ecosystem_shift OpenAI’s Executive Shuffle Gets More Serious — Fidji Simo Steps Back and the AGI Org Suddenly Looks Less Settled

infrastructure The Power Bill Comes Due — Google-Backed AI Capacity Is Being Fed by a Giant Gas Plant

policy Utah Just Let an AI Prescribe Psychiatric Drugs — The Permission Frontier Moved Again

security_advisory The AI Supply Chain Just Sprang a Leak — Meta Halts Work With Mercor After a Breach Threatens Training Secrets

ecosystem_shift Compute Has Entered Geopolitics for Real — Iran Threatens Stargate-Linked AI Infrastructure in Abu Dhabi

ecosystem_shift OpenAI Buys TBPN and Makes the Quiet Part Loud — Distribution and Narrative Control Are Now Product Strategy

ecosystem_shift Anthropic Starts Closing the Harness Loophole — Claude Subscriptions No Longer Ride Free Inside OpenClaw and Other Third-Party Clients

ecosystem_shift OpenAI's Product Chief Steps Back for Medical Leave as the Company Quietly Reshuffles Its Power Centers

ecosystem_shift The AI Boom Hits the Grid — Up to Half of Planned U.S. Data Center Builds Are Slipping or Dying on the Vine

security_advisory Claude Code Helps Surface a Linux Kernel Bug That Lived for 23 Years — Security Research Just Got Faster and Weirder

security_advisory The Claude Code Leak Just Got Meaner — Malware Is Piggybacking on Reposted Source Archives

ecosystem_shift OpenAI’s ‘Super App’ Vision Is Not a Feature Roadmap — It’s a Bid to Own the Entire Agent Interface

ecosystem_shift Anthropic Kills Claude Subscriptions on Third-Party Harnesses — OpenClaw, Others Cut Off Tomorrow at Noon PT

model_release Netflix Drops VOID — Their First Public Model, Apache 2.0, Removes Objects and Their Physical Interactions from Video

ecosystem_shift Half of Planned US Data Center Builds Delayed or Canceled — Power Infrastructure and China Parts Shortages

research_paper Anthropic Finds 171 Emotion Vectors Inside Claude — Not Metaphors, Actual Neuron Patterns Steering Behavior

ecosystem_shift Altman Shuts Down Sora, Says 'Something Very Big and Important Is About to Happen' with Next-Gen Models

ecosystem_shift OpenAI Acquires TBPN — 209 HN Points as Deal Details Emerge

policy Pro-AI SuperPAC Drops $100 Million on US Midterm Elections as Public Backlash Grows

ecosystem_shift Bay Area Therapists Say AI Workers Are in Crisis — Anxiety, Burnout, and Existential Dread About What They're Building

model_release Qwen3.6-Plus Drops — Alibaba's Latest Frontier Model Ships with Open Weights Coming

breaking_news Claude Code's Full Source Leaked via npm Source Maps — 500K Lines of TypeScript Laid Bare, Working Executable Rebuilt in Hours

ecosystem_shift OpenAI Closes Record $122 Billion Funding Round — Valued at $852 Billion as IPO Anticipation Grows

policy Microsoft Copilot TOS Now Says It's 'For Entertainment Purposes Only' — And Nobody's Laughing

ecosystem_shift ChatGPT Quality Collapse Goes Viral — 6,200 Upvotes on 'WTF' Post as GPT-5.4 Complaints Flood Reddit

ecosystem_shift 'The Era of Human Coding Is Over' — 2,989 Upvotes as AI Coding Discourse Reaches Fever Pitch

ecosystem_shift Nicolas Carlini: Claude Is a Better Security Researcher Than Me — Made $3.7M in Bug Bounties

ecosystem_shift Dario Amodei Compares Altman to Stalin, Calls Brockman's $25M Trump Donation 'Evil' — WSJ Reveals Decade-Long AI Feud

model_release Gemma 4 Specs Leak on Twitter — Google's Next Open-Weight Frontier Model Reportedly Incoming

ecosystem_shift Anthropic May Have Had an Architectural Breakthrough — Mythos Reportedly 'Far Above Internal Expectations'

breaking_news Anthropic Left 'Mythos' Model Details in an Unsecured Public CMS — Fortune Calls It a 'Step Change' with Unprecedented Cyber Risks

price_change Anthropic Tightens Peak-Hour Session Limits — Pro Users Report 2 Prompts Burning Entire 5-Hour Windows

ecosystem_shift OpenAI Shelves Adult Mode, Kills Sora, Cancels Stargate — Consolidates Everything Around 'Spud'

research_paper ARC-AGI-3 Drops and Every Frontier Model Scores Below 1% — Chollet's Hardest Benchmark Yet

policy Sanders and AOC Introduce Data Center Moratorium Act — Construction Freeze Until AI Safeguards Exist

policy Federal Judge: Pentagon's Anthropic Blacklisting 'Looks Like Punishment' for AI Safety Views

security_advisory LiteLLM Supply Chain Attack: PyPI Versions 1.82.7 & 1.82.8 Ship Credential-Stealing Backdoor — Thousands Potentially Compromised

ecosystem_shift OpenAI Kills Sora — App, API, Disney Deal All Discontinued as Open-Source Video Gen Eats the Market

ecosystem_shift Microsoft Threatens to Sue OpenAI Over $50B AWS Deal — 'Frontier' Product Allegedly Violates API Routing Clause

ecosystem_shift Anthropic Ships Dispatch — Claude Now Uses Your Computer While You Watch from Your Phone

breaking_news Epoch Confirms GPT-5.4 Pro Solved a Frontier Math Open Problem — First AI to Crack Ramsey Hypergraphs

ecosystem_shift Jensen Huang Tells Lex Fridman 'I Think We've Achieved AGI' — Then Defines Token Factories as the New Economy

policy Pentagon Selects Palantir AI as Core US Military System — Sovereign AI Stack Takes Shape

policy Neil DeGrasse Tyson Calls for International Treaty Banning Superintelligence — 'That Branch of AI Is Lethal'

research_paper OpenAI Research: Models 'Go Insane' When Given Repetitive Tasks from Automated Users

research_paper Anthropic's Own Research Shows AI Coding Tools Are Making Developers Measurably Worse

ecosystem_shift Karpathy: 'I Haven't Written a Line of Code Since December' — 16 Hours a Day Directing Agents

ecosystem_shift DeepSeek Core Researcher Daya Guo Has Reportedly Resigned

framework_release vLLM v0.18.0: gRPC Serving, 445 Commits, and Known FP8 Accuracy Issue on B200

framework_update llama.cpp b8460: Fix for Silent Prompt Corruption Bug

ecosystem_shift Cursor Composer 2 Built on Kimi K2.5 — Attribution Saga Ends in Awkward Reconciliation

ecosystem_shift OpenAI Acquires Astral — uv, Ruff, and ty Now Under the Codex Umbrella

research_paper Nemotron-Cascade 2: NVIDIA's 30B MoE Hits Gold Medal on IMO, IOI, and ICPC with Only 3B Active Params

framework_update langchain 1.2.13 + langchain-anthropic 1.4.0: LangSmith Metadata and Prompt Caching Middleware

ecosystem_shift 73% of Enterprise AI Spend Now Flows to Anthropic — Axios Reports Revenue Shift

framework_update OpenAI Agents SDK v0.12.5: MCP Auth Exposure, Nested Agent Recovery, Session Retry Fixes

model_release Mamba 3: Together AI Ships State Space Model Optimized for Inference

model_release GPT-5.4 Mini and Nano Released — Small Models for Coding and High-Volume API Work

tool_release Unsloth Studio Launches — Apache-Licensed Open-Source UI for Training and Running LLMs

tool_release Claude Cowork Gets Remote Access — Control Your Desktop Agent from Your Phone

ecosystem_shift NVIDIA GTC 2026: DLSS 5 Introduces Generative AI Rendering, Vera Rubin Projections Hit $1T

model_release Mistral Small 4 Released — Compact Model with Strong Tool-Use and Structured Output

security_advisory Teens Sue xAI Over Grok-Generated CSAM — Minors Allegedly 'Undressed' by AI

framework_update Nvidia Removes Rug-Pull Clauses from Nemotron 3 Super License

framework_update OpenAI Agents SDK v0.12.0-0.12.3: Four Releases in Five Days

technique Simon Willison Publishes Definitive Agentic Engineering Guide

model_release NVIDIA Nemotron 3 Super: 120B open model optimized for agentic reasoning

tool_release Ollama v0.18.0: Cloud model support and 2x faster Kimi-K2.5

framework_release A2A Protocol v1.0.0: Google's Agent2Agent hits stable with breaking changes

policy Americans Prefer Banning AI Over Unregulated Development by 4-to-1 Margin

policy US Military Using Claude AI to Plan Iran Air Attacks Despite Anthropic-Pentagon Clashes

api_change 1M Context Window GA for Claude Opus 4.6 and Sonnet 4.6 — No Long-Context Premium

ecosystem_shift Nvidia Commits $26B to Building Open-Weight AI Models

tool_release Perplexity 'Personal Computer' — Dedicated Mac as Always-On AI Agent

tool_release Gemini Task Automation Ships on Samsung S26 and Pixel 10

model_release NVIDIA Nemotron 3 Super: 120B Hybrid Mamba-Transformer MoE with 12B Active Parameters

ecosystem_shift NVIDIA Commits $26 Billion to Open-Weight AI Model Development

tool_release llama.cpp Ships Real Reasoning Budget Support via Sampler Mechanism

tool_release Claude Code Review Launches — Multi-Agent PR Reviews for Team/Enterprise

model_release GPT-5.4 Thinking and Pro Rolling Out — OpenAI's Unified Frontier Model

security_advisory Clinejection: Prompt Injection Compromised Cline's Production Releases via Issue Triager

industry_movement Anthropic Sues Trump Administration Over Pentagon Blacklisting

infrastructure OpenAI Walks Away from Oracle Stargate Data Center Expansion

industry_movement Yann LeCun Launches AMI Labs with $1B Seed — Bets Against LLMs

security_advisory Clinejection: Cline's Production Releases Compromised via Prompt Injection in Issue Triager

tool_release Agent Safehouse: macOS-Native Sandboxing for Local Agents

model_release GPT-5.4 Released: Two New API Models from OpenAI