The week in AI, Nov 9 – Nov 15, 2026
4 daily briefs, 38 stories and 29 dedicated articles, with every source linked. Start with the day you missed.
- Nov 12, 2026: Nvidia's board weighs removing Huang as OpenAI reports a $10.9 billion quarterly loss
- Nov 11, 2026: Anthropic prices its IPO at up to $1.4 trillion as Google opens Gemini 4 Carbon
- Nov 10, 2026: Frontier tokens get cheaper from both ends as SoftBank posts a $7 billion loss
- Nov 9, 2026: Nvidia's board investigates its CEO as agents run both sides of the breach
Dedicated coverage this week
Security
Railway's post-mortem: a build-sandbox escape, one over-trusted credential and $740,000 of fraudulent AI charges
The clearest public account yet of how a platform breach becomes LLMjacking, with the exact controls that limited the damage and the ones that did not…
Operation Snowball: a jailbroken DeepSeek agent ran a two-week telecom intrusion with a part-time operator, and every exploit was a known CVE
Malbear Labs recovered 70 session transcripts from an attacker's open directory; what they show about cost, autonomy and refusal training should chang…
Fireworks AI's second breach in a month: what inference customers should rotate, review and ask for now
An October 24 intrusion through a cloud provider reached billing data, secrets, datasets and partial model weights, weeks after stolen internal creden…
Claude Code's 'don't ask again' leaked 2,539 secrets to GitHub. A fix-it checklist
The convenience button that stops permission prompts also wrote passwords into a project file that thousands of developers committed. Here is how it h…
Health & medicine
Gemini for healthcare: Google's agent reads Epic, Medicare rules and payer systems for seven launch customers; what to ask before you pilot it
Northwell has 400 people fighting insurance denials and wins 90% of them; Google wants both sides of that fight running on its agent. The economics, t…
Oura pays $625 million for Counsel Health: a 70-person AI primary-care startup is now worth 17 times what it raised
A smart-ring maker that could not get its own IPO away is paying up for a chat-first clinic with physicians in the loop; the price says where consumer…
Abridge cuts staff at a $5.3 billion valuation: what it says about the AI scribe market, and what health systems should ask
The best-funded company in ambient clinical documentation is flattening its organisation as the category commoditises; buyers mid-contract should get …
The PATIENT AI Act draft: a third-party FDA review path and a Medicare pathway, comments due December 1
Senator Cassidy's discussion draft is the first serious attempt to legislate how clinical AI gets reviewed and paid for. FDA lawyers say the review ti…
Finance & markets
Nvidia's board weighs removing Jensen Huang: what an 8.6% after-hours drop and a $200 price target are pricing in
Five days before earnings, the world's second-most-valuable company may replace the only chief executive it has ever had; here is how to read the stoc…
Anthropic's IPO range is $70–78: what the amended S-1 says about dilution, lock-ups and the $26 billion partner placement
The largest technology listing ever attempted now has a price, a share count and a list of suppliers buying in; here is how to read the numbers before…
SoftBank's $7 billion loss lands on the day Anthropic starts selling a $1.1–1.3 trillion IPO
The largest outside holder of OpenAI just reported its first quarterly loss in two years, and the valuation range for Anthropic's listing has fallen f…
Anthropic's reported $3.4B generator buy makes power a capital expense for AI labs
Taking over 1.7 gigawatts of Caterpillar orders is the biggest known power-equipment purchase by an AI company. It tells you how the labs now think ab…
Robotics
Skild AI's Steel runs robot simulation up to 13x faster than PhysX on mixed batches and uses 62% less memory; here is what the numbers cover
A $14 billion startup contributed its rigid-body solver to Nvidia's open-source Newton engine a week after Nvidia shipped Isaac Lab 3.0; the benchmark…
Nvidia's GR00T 2 claims 71% one-shot success on 50 tasks. Here is what that number does and does not say
A new open world-action model trained on 100,000 hours of human video, a simulator that agents can configure themselves, and a benchmark claim that ne…
Agility Robotics is the first public humanoid company. Its numbers explain why that is hard
$174,000 of revenue, a $362 million half-year loss and $300 million of orders for a robot that ships in 2027: AGLT is a bet on throughput targets nobo…
Models & research
Gemini 4 Carbon is open to every paying developer at $2 / $10 until November 30: what to test before the price doubles
Google's strongest model left its cyber-only gate after two weeks, matched Argon's price for 19 days and published safety numbers that are worth readi…
Claude Fable 5.2 cuts the top tier to $8 / $40: what actually changes for buyers
Anthropic's flagship is 20% cheaper and, on its own numbers, far better at agentic coding, but the company still tells most customers to stay on Opus …
Inkling 2 Small: a 594B-parameter model trained in 68 hours, priced at $0.30 per million tokens
Thinking Machines' second-generation model is a case study in how cheap sparse pretraining has become, and a direct price attack on the mid-tier where…
Policy & regulation
New York opens RAISE Act registration: $500 million revenue, 10^26 FLOPs, 72-hour incident reports and a January 1 deadline
The first US state registry of frontier AI developers is now taking filings; here is who is covered, what they must publish, what the penalties are an…
Australia's AI safety regulator: mandatory incident reporting, penalties and model evaluation after the Medicare breach
The first Western government to answer a specific AI-agent incident with a statutory regulator has set a template other capitals will copy; the detail…
Enterprise & business
United Airlines wrote AI agents into its contract of carriage as a prohibited practice: what the text says and what retailers should take from it
Supervision is no defence, searching is still allowed and United can cancel without notice; the first airline ban on agents is a template other mercha…
OpenAI's $10 billion quarter came with a $10.9 billion loss: what the leaked numbers say next to Anthropic's S-1
Revenue up 50% in a quarter, enterprise now bigger than consumer, and a loss of $1.09 per dollar of sales; the figures explain the Cursor cutoff, the …
Claude Sheets, Whiteboards and Meetings: Anthropic now sells an office suite, and here is what IT buyers should ask before the renewal
The day before its roadshow, Anthropic turned Claude from a chat window into a workspace with spreadsheets, shared boards, embedded partner apps and a…
Google's four TPUs in one year: what 8tx and 8i Max mean for anyone negotiating with Nvidia
Confidential compute for fine-tuning and a doubled-up inference chip extend the eighth-generation line, and the software around it is now the real com…
Nvidia's board is investigating Jensen Huang. Here is what is actually at stake
An independent review, an admitted affair and a November 18 congressional briefing turn a personal scandal into a governance test for the company the …
Developer tools
Gemini Decision 1: 1,000 questions per call, 150 ms, $0.10 per million input tokens and no output bill
Google and OpenAI now both sell decisions as a product separate from text; here is what the new API does, where it beats a Flash model, and the three …
AlphaProof Code verified Signal's protocol code with 6.5 million lines of Lean in four weeks: what AI-written proofs mean for the software you depend on
DeepMind's five-person team used agents on Gemini 4 Carbon and Claude Opus 5.5 to prove security properties of libsignal at the machine-code level; th…
Muse Ember at $0.10 and LongCat-2.5 under MIT: the budget tier is now where agent costs are decided
Meta's smallest model matches Claude Haiku 5.5 on the Artificial Analysis index at a third of the cost per task, and Meituan has handed developers a 1…
OpenAI's GPT-5.6 Sol discount ends November 30. Here is the price ladder to move to
Two vendors' promotional rates expire within a day of each other. OpenAI's pricing page now shows four GPT-6 tiers from $0.10 to $10 per million token…
Model releases this week
Release tracker →Thursday, November 12, 2026
Full brief →Nvidia's board weighs removing Huang as OpenAI reports a $10.9 billion quarterly loss
Nvidia's board met to decide Huang's future; shares fell 8.6% after hours and 6% premarket
The Wall Street Journal reported on November 11 that Nvidia's board was preparing to remove Jensen Huang as chief executive, and Reuters reported that…
OpenAI's revenue reached $10 billion in Q3 while its adjusted operating loss grew to $10.9 billion
Internal documents reviewed by the Wall Street Journal and reported on November 11 show OpenAI's third-quarter revenue at about $10 billion, up roughl…
Railway's post-mortem: a build-sandbox escape led to $740,000 in fraudulent AI charges on stolen keys
Railway published its full incident report on November 10 for the October 20–21 breach. An attacker exploited a previously unknown flaw to partially e…
Google's Gemini Decision 1 scores up to 1,000 questions per call at $0.10 per million input tokens, no output charge
Google launched the Gemini Decisions API on November 11, powered by a model called gemini-decision-1 that scores a fixed set of options in a single fo…
Google Cloud puts a Gemini agent inside Epic, billing and insurer systems for Northwell, HCA and Optum
Google Cloud launched Gemini for healthcare on November 10 as a US preview, an agent that reads records from Epic and Oracle Health, billing platforms…
New York opens RAISE Act registration: frontier labs over $500 million in revenue must file by January 1
Governor Kathy Hochul announced on November 10 that registration is open for large frontier developers under the RAISE Act, which takes effect on Janu…
United Airlines bans AI agents from booking in its contract of carriage, supervised or not
United revised its contract of carriage on November 11 to define an AI agent as any assistant, standalone agent or AI-enabled browser that discovers, …
OpenAI's models leave Cursor today, 75 days after the SpaceX takeover triggered the cutoff
November 12 is the date OpenAI set in August for ending its supply of models to Cursor, after SpaceX closed its $60 billion all-stock acquisition of t…
Meta's infrastructure chief steps down as a $135–145 billion capex year hands over to an ex-AWS executive
Santosh Janardhan, Meta's head of infrastructure and co-head of engineering since 2022 and an employee since 2009, told staff in a November 11 memo th…
Wednesday, November 11, 2026
Full brief →Anthropic prices its IPO at up to $1.4 trillion as Google opens Gemini 4 Carbon
Anthropic sets its IPO range at $70–78, raising up to $64 billion at a $1.4 trillion valuation
Anthropic launched its roadshow on November 10 with an amended S-1 offering 825 million Class A shares at $70 to $78 under the Nasdaq ticker ANT, whic…
Gemini 4 Carbon opens to all paying developers at $2 / $10, doubling to $4 / $20 on December 1
Google made Gemini 4 Carbon generally available on November 10 through the Gemini API, AI Studio, Gemini for business, Android Studio and Antigravity,…
Anthropic ships Claude Sheets, Whiteboards and interactive connectors, with Meetings to follow
A day before its roadshow, Anthropic released Claude Sheets, a collaborative spreadsheet where Claude writes formulas and logs every change, Claude Wh…
A jailbroken DeepSeek agent ran a two-week intrusion into a European telecom and 13 of its tenants
Malbear Labs published a reconstruction on November 8 of what it calls Operation Snowball, in which an operator pointed a custom agent framework runni…
35,578 Grok Build apps were open to anyone with a link, including 24 built by SpaceXSI and X staff
Security firm Claroc reported on November 10 that 35,578 of the 37,571 live apps it enumerated from Grok's chat-based Build mode, about 95%, could be …
Notion confirms AI chat transcripts from under 1% of workspaces leaked through the Fireworks breach
Notion's advisory, updated November 10, says Fireworks AI told it on November 2 that attackers had reached what was first described as synthetic test …
Oura buys AI primary-care startup Counsel Health for $625 million after shelving its IPO
Smart-ring maker Oura agreed on November 10 to acquire Counsel Health, a chat-first medical service that pairs an AI assistant with licensed physician…
Skild AI open-sources Steel, a Newton solver it says runs robot simulation up to 13x faster than PhysX
Skild AI, the $14 billion robot-foundation-model startup backed by SoftBank and Nvidia, released Steel on November 9 as a solver inside Nvidia's open-…
DeepMind's AlphaProof Code wrote 6.5 million lines of proof for Signal's encryption in four weeks
Google DeepMind said on November 10 that AlphaProof Code, a system pairing five researchers with agents running on Gemini 4 Carbon and Claude Opus 5.5…
OpenAI turns ChatGPT into a business directory with claimable profiles and sponsored Business Agents
OpenAI launched Businesses in ChatGPT in beta on November 10: a Business Hub where companies claim a profile, upload product feeds with prices and sto…
Tuesday, November 10, 2026
Full brief →Frontier tokens get cheaper from both ends as SoftBank posts a $7 billion loss
Claude Fable 5.2 and Mythos 5.2 arrive at $8 / $40, 20% below the 5.1 price
Anthropic released Claude Fable 5.2 and Claude Mythos 5.2 on November 9 at $8 per million input tokens and $40 output, down from $10 / $50 for the 5.1…
SoftBank swings to a ¥1.1 trillion loss as Anthropic's IPO roadshow opens at a reported $1.1–1.3 trillion
SoftBank Group reported a net loss of about ¥1.1 trillion ($7 billion) for July–September on November 10, against a ¥2.5 trillion profit a year earlie…
Australia will set up an AI safety regulator with penalty powers and mandatory incident reporting
Prime Minister Anthony Albanese said on November 10 that AI companies operating in Australia will face mandatory safeguards enforced by a new AI safet…
Meta's Muse Ember at $0.10 / $0.50 and Meituan's MIT-licensed LongCat-2.5 reset the budget tier
Meta released Muse Ember, its smallest multimodal model, aimed at high-volume agent sub-tasks such as tool calls, routing and extraction, priced at $0…
Google adds TPU 8tx for confidential compute and TPU 8i Max for long context, its fourth TPU system this year
Google announced two more eighth-generation TPU systems on November 9: TPU 8tx, built on the same silicon as the training-focused 8t and paired with f…
Nvidia opens GR00T 2 and Isaac Lab 3.0 at CoRL; 71% one-shot success on an unnamed 50-task suite
At the Conference on Robot Learning in Austin on November 9, Nvidia introduced GR00T 2, a world action model built on Cosmos and trained on robot data…
Fireworks AI tells customers of a second breach in a month, reaching billing data and some model weights
Fireworks AI notified customers on November 9 that an intruder entered development machines on October 24 through a cloud provider and reached billing…
Abridge, the $5.3 billion AI scribe, confirms layoffs in an 'organisation-wide realignment'
Abridge confirmed on November 9 that it has laid off staff as part of what it called an organisation-wide realignment to flatten the company, offering…
ChatGPT Enterprise lands in AWS Marketplace, letting buyers draw down cloud commitments
OpenAI listed ChatGPT Enterprise in the AWS Marketplace on November 9, so organisations can buy it through their existing AWS billing relationship and…
Monday, November 9, 2026
Full brief →Nvidia's board investigates its CEO as agents run both sides of the breach
Nvidia's board opens an independent probe after Huang admits an affair with the alleged Megaspeed informant
Nvidia's board said on November 8 that independent directors and outside counsel will review CEO Jensen Huang's relationship with Alice Huang, a Megas…
Anthropic reportedly paid $3.4 billion for 1.7 gigawatts of gas generators
Anthropic took over OnePWR Solutions' orders for roughly 1.7 gigawatts of Caterpillar gas-fired generators in a deal signed in late September and wort…
Claude Code's 'don't ask again' button put 2,539 secrets in public GitHub repos
Security firm Apiiro found that when a Claude Code user approves a shell command with 'Yes, and don't ask again', the exact command is saved to .claud…
Check Point: autonomous ARTEX agents hit 200+ organisations in 14 countries since October 17
An AI-driven intrusion campaign using the ARTEX penetration-testing framework has targeted more than 200 organisations across 14 countries since Octob…
DeepMind says AI agents found two weaknesses in 1024-bit RSA; 2048-bit keys unaffected
Google DeepMind said on November 6 that supervised AI agents produced two working papers describing weaknesses in legacy 1024-bit RSA: a signature-for…
Thinking Machines ships Inkling 2 Small: 594B-parameter MoE, 17B active, weights promised under Apache 2.0
Thinking Machines Lab released Inkling 2 Small on November 5, a sparse mixture-of-experts model with 594 billion total and 17 billion active parameter…
OpenAI's GPT-5.6 Sol discount ends November 30; the real price ladder is now GPT-6.1 at $2 / $10
OpenAI's pricing page confirms that promotional pricing on GPT-5.6 Sol ($4 input / $20 output per million tokens) runs only until November 30, after w…
Physical AI goes public: Agility Robotics trades on Nasdaq as Axon puts $1.2 billion into Skydio
Agility Robotics became the first pure-play humanoid company on a US exchange on November 5, trading as AGLT after closing its merger with Churchill C…
Cassidy's PATIENT AI Act draft would add a third-party FDA review path for health AI; comments due December 1
Senate HELP Committee chair Bill Cassidy's discussion draft, released October 29 with stakeholder comments requested by December 1, proposes a volunta…
Anthropic tells enterprise customers an outside security firm viewed account data; Korean regulator reviews ~100 accounts
Anthropic has emailed enterprise customers that an outside cybersecurity firm accessed internal systems in late October, exposing account and organisa…
Get the brief in your inbox
One email every morning: the stories that mattered in AI, each with a one-line "why it matters" for your role. Free, no ads, unsubscribe any time.
Published daily by AI Briefing. We never share your address.