# Times of AI

[https://timesof.ai/](https://timesof.ai/)

Hedged briefs and analysis on the AI economy.

## Latest

- [Google's First Gemini Glasses Go on Preorder at $395, Betting That Privacy Limits Will Sell an AI Camera on Your Face](https://timesof.ai/2026/11/google-samsung-gemini-intelligent-eyewear-preorders-395-december-10) — Google's Gemini Intelligent Eyewear with Samsung, Warby Parker and Gentle Monster is on preorder from $395, shipping Dec 10 in the US and South Korea.
- [Nscale Says Its Anthropic Contracts Pushed It Into Chip Orders It Cannot Cancel, and It Needs New Financing to Pay for Them](https://timesof.ai/2026/11/nscale-amended-s1-needs-financing-anthropic-monarch-purchase-orders) — Nscale's amended S-1 says some chip orders tied to its $44.6B Anthropic deal can't be cancelled and it needs new financing for 2027. Nine-month loss $1.69B.
- [Reuters Finds $3.2 Billion of Chinese AI Projects Designed Around Nvidia Chips the US Restricts, a Bet That Now Rests on Washington](https://timesof.ai/2026/11/reuters-3-2b-chinese-ai-projects-built-on-restricted-nvidia-chips) — Reuters found 23B yuan of Chinese AI computing projects planned around restricted Nvidia H100, H200 and H800 chips. The total includes buildings and power.
- [Six Chinese Open-Weight Models Carry a Blackwell Fingerprint, an Independent Weight Analysis Finds, but It Cannot Say Where the Chips Were](https://timesof.ai/2026/11/blackwell-fingerprints-deepseek-moonshot-minimax-ant-open-weights-unemployment-labs) — An Unemployment Labs analysis says DeepSeek, Moonshot, MiniMax and Ant Group models were initialized on Nvidia Blackwell GPUs. Location and legality unproven.
- [Mirendil Raises $1 Billion to Sell the Idea That One Researcher With Agents Can Pretrain a Model](https://timesof.ai/2026/11/mirendil-1b-series-a-kleiner-clearlake-agent-run-ai-research) — AI lab Mirendil raised a $1B Series A led by Kleiner Perkins and Clearlake to sell agent-run AI research. Valuation undisclosed; claims unverified.
- [Anthropic Lets Companies Run Claude's Agent Service Behind Their Own Firewall, but Claude Itself Still Runs Outside It](https://timesof.ai/2026/11/anthropic-self-hosted-claude-managed-agents-private-beta) — Self-hosted Claude Managed Agents keeps agents, credentials and history in your environment; model calls still go to Anthropic or a cloud provider.
- [OpenAI Turns Its Cyber Models Into Products: Daybreak Red Team Attacks Your Web Apps, Defense Factory Writes the Patches](https://timesof.ai/2026/11/openai-daybreak-red-team-defense-factory-agent-pen-testing) — OpenAI's Daybreak Red Team (early access) pen-tests web apps with agents; Defense Factory drafts patches for engineer review. No pricing or outside results yet.
- [Google's Cheaper Gemini 4 Hydrogen Scores 51 on an Independent Index, Though at Its Top Setting It Still Ranks 16th](https://timesof.ai/2026/11/google-gemini-4-hydrogen-everyday-agent-model-1-50-7-50) — Google's Gemini 4 Hydrogen targets coding and agent work at $1.50/$7.50 per million tokens. Artificial Analysis scores it 51 (high) and 54 (xhigh).
- [Nonprofits Ask a California Court to Freeze Frontier AI, Using the Labs' Own Risk Warnings as Their Evidence](https://timesof.ai/2026/11/nonprofits-sue-to-freeze-frontier-ai-anthropic-google-meta-openai-xai) — Public Citizen and three other groups sued Anthropic, Google, Meta, OpenAI and xAI in San Francisco, seeking to freeze more capable AI under California law.
- [Databricks Turns Genie Into an Always-On Enterprise Agent, but the Autonomous Part Is Still Only Headed for Preview](https://timesof.ai/2026/11/databricks-always-on-genie-agent-ontology-ga) — Databricks' second-generation Genie runs in the background with sub-agents and an inbox. Genie Ontology is GA; the always-on agent is coming to preview.
- [MiniMax's Open-Weight H3-Image Makes 4K Pictures in Seconds, Under a Licence That Shuts Out the US, EU, UK and Korea](https://timesof.ai/2026/11/minimax-h3-image-open-weights-4k-license-excludes-us-eu-uk-korea) — MiniMax's open-weight H3-Image (33.1B DiT, six steps) claims 4K in under 10s. Licence excludes the US, EU, UK and Korea; $1M revenue cap.
- [United Airlines Bars Unauthorized AI Agents From Booking and Check-In, and Extends the Ban to Travel Agencies](https://timesof.ai/2026/11/united-airlines-restricts-unauthorized-ai-agents-bookings) — United bars unauthorized AI agents from booking, check-in and account changes, including via travel agencies, citing security and FAA/DOT rules.
- [SoftBank Leads $500 Million Round Valuing Agile Robots at $8.5 Billion, Less Than the Round Discussed in June](https://timesof.ai/2026/11/agile-robots-500m-series-f-8-5b-valuation-softbank-nvidia) — Munich robotics firm Agile Robots closed a $500M Series F at $8.5B led by SoftBank, with Nvidia and Hitachi, to expand in the US and build capacity.
- [Chinese AI Firms Rush to Lock In Overseas Nvidia Capacity Before Washington Closes the Cloud Loophole, Reuters Reports](https://timesof.ai/2026/11/chinese-ai-firms-rent-overseas-nvidia-capacity-remote-access-reuters) — Reuters: Chinese AI firms are trying to reserve overseas Nvidia compute, fearing allegations around former CEO Jensen Huang speed the Remote Access Security Act. Firms unnamed.
- [Helsing Launches Clarity, Agent-Based Military Intelligence Software That Prepares Targets for Human Approval](https://timesof.ai/2026/11/helsing-clarity-agentic-military-intelligence-platform) — Helsing's Clarity uses hundreds of AI agents to fuse battlefield data and prepare targets for human approval. Reuters: a Palantir Maven rival; no customers named.
- [Runway Opens the Weights of Its Video-Trained Robot Model, the Top World Action Model on Nvidia's Leaderboard but Fifth Overall](https://timesof.ai/2026/11/runway-praxis-1-open-weights-robot-model-robolab) — Runway's 27B Praxis-1 world action model is open-weight. It leads world action models on RoboLab-120 at 46.2% but ranks fifth overall on the leaderboard.
- [Torus Raises $500 Million to Put AI Inference Where Spare Grid Power Already Sits, in Units It Says Go Live in 12 to 16 Weeks](https://timesof.ai/2026/11/torus-500m-foundry-modular-inference-spare-grid-power) — Magnetar-led $500M backs Torus Foundry: 1-20 MW inference units on existing grid connections, live in 12-16 weeks, per Torus. Hydra Host is a first customer.
- [Meta Hands Its AI Data Center Build-Out to a Former AWS Executive as Infrastructure Chief Santosh Janardhan Steps Down](https://timesof.ai/2026/11/meta-infrastructure-chief-janardhan-steps-down-dave-brown) — Meta's Santosh Janardhan will step down in 2027; ex-AWS SVP Dave Brown takes over infrastructure under Meta Compute as capex runs $135B-$145B this year.
- [A Blacklisted Chinese Defence University Bought Nvidia H200s Through a Local Supplier, Reuters Finds in Procurement Records](https://timesof.ai/2026/11/nvidia-h200-northwestern-polytechnical-entity-list-reuters) — Reuters: Entity-listed Northwestern Polytechnical bought two H200s via a local supplier and rented H200 servers, showing export-control gaps. Nvidia did not respond.
- [DeepMind's AlphaGenome 2 Narrows DNA Predictions From Whole Tissues to Single Cell Types](https://timesof.ai/2026/11/deepmind-alphagenome-2-cell-type-dna-variant-predictions) — AlphaGenome 2 predicts how DNA variants act in specific cell types; DeepMind says it finds nearly twice as many expression-changing variants. Research use only.
- [Google's New Gemini Model Doesn't Write Answers, It Picks Them: Decision 1 Scores Up to 1,000 Questions for $0.10 per Million Tokens](https://timesof.ai/2026/11/google-gemini-decisions-api-decision-1-closed-set-choices) — Google's Gemini Decisions API scores yes/no, choice and rubric questions with confidence scores at $0.10/M input tokens. An independent index ranks it 9th of 18.
- [AWS Ships a Free Local Copy of Its Cloud That It Says No Human Wrote or Reviewed](https://timesof.ai/2026/11/aws-emulator-built-entirely-by-ai-agents-cc0) — AWS Emulator mimics 200+ AWS services locally for developers and coding agents. AWS says agents wrote all the code with no human review, gated by agent checks.
- [One GitHub Issue Could Have Led to a Poisoned Codex Release, Researchers Say; OpenAI Has Fixed the Bug](https://timesof.ai/2026/11/openai-codex-prompt-injection-chain-npm-publishing-depi) — Depi says a prompt injection in a Codex GitHub workflow could be chained to npm publishing credentials. OpenAI fixed it; no malicious release reported.
- [Oura Agrees to Buy Counsel Health for $625 Million, Betting That a Smart Ring Plus a Medical AI and On-Call Doctors Is a Care Business](https://timesof.ai/2026/11/oura-counsel-health-625m-ai-native-primary-care) — Oura agreed to buy Counsel Health, which pairs medical AI with on-call physicians, for $625M cash. Oura reports preliminary Q4 revenue of $533M, up 153%.