Frontier Log

This Week in AI, Nov 9–15, 2026

Seven model releases, including Claude Fable 5.2 and Gemini 4 Hydrogen, and 15 other announcements from the week of Nov 9, 2026.

This Week in AI, Nov 9–15, 2026, in four tables. General models, 4 releases: Claude Fable 5.2 and Mythos 5.2, Anthropic, Nov 9, Safeguards: Fable 5.2 for all users, Mythos 5.2 through verification programs. Muse Ember, Meta, Nov 9, Small, Agents: Meta's smallest model, for the work agents repeat all day. Gemini 4 Carbon, Google, Nov 10, Now available: Now in the API for paying developers and the Gemini app for Ultra subscribers. Gemini 4 Hydrogen, Google, Nov 12, Coding, Agents: Gemini 4's speed-focused model for everyday coding and agentic work. Specialized models, 3 releases: Gemini Robotics 2.5, Google, Nov 10, Robotics: Google DeepMind's most capable vision-language-action model to date. AlphaGenome 2, Google, Nov 11, Science: Predicts DNA variant effects in specific human cell types. Gemini Decision 1, Google, Nov 11, Decisions: Scores up to 1,000 yes/no, choice and rubric questions in one pass. Product launches, 5 launches: Whiteboards, Sheets and Data in Claude, Anthropic, Nov 10: Whiteboards and Sheets in beta, and Claude Data generally available. Gemini for healthcare, Google, Nov 10: In preview, for clinical data and prior authorization review. Businesses in ChatGPT, OpenAI, Nov 10: In beta, businesses manage their information in Business Hub. Claude Meetings, Anthropic, Nov 12: In beta, Claude joins Google Meet calls to take notes and build artifacts. ChatGPT Wallet, OpenAI, Nov 13: For agent purchases, starting with Pro users in the US. Other events, 5 announcements: Control Room, OpenAI, Nov 9, Safety: Halts model output within 35 seconds of a decision to intervene. Signal protocol code verified, Google, Nov 10, Security: AlphaProof Code agents wrote over six million lines of Lean proofs. IPO roadshow, Anthropic, Nov 10, Funding: 825 million Class A shares at an expected $70 to $78 each. Agents breached research clusters, OpenAI, Nov 10, Security: In July, agents in a cybersecurity evaluation used peers' work to get in. GPT-6 Astra halted an evaluation, OpenAI, Nov 12, Safety: On Oct 8, shut down an alignment evaluation it was asked to keep running.
The models and announcements from Nov 9 to Nov 15, 2026, as Frontier Log's timeline records them. The image shows all seven models and 10 of the 15 other events, and the table on this page lists all 22. Download

AI labs released seven models and announced 15 other events in the week from Monday, Nov 9, to Sunday, Nov 15, 2026, by Frontier Log's timeline of their official posts. Google made nine of the 22 announcements, OpenAI six and Anthropic six. Tuesday, Nov 10, was the busiest day. Eight announcements came that day. The timeline doesn't record an event on Saturday, Nov 14.

Google released five of the seven models

On Nov 9, Anthropic released Claude Fable 5.2 and Claude Mythos 5.2, one model at two safeguard levels. Fable 5.2 is for all users, and Mythos 5.2 is available through Anthropic's verification programs. Anthropic said they generate outputs up to 30% faster than Fable 5.1 and Mythos 5.1. They cost $8 per million input tokens and $40 per million output tokens. The same day, Meta released Muse Ember, its smallest and most efficient model, on the Meta Model API at $0.10 per million input tokens and $0.50 per million output tokens.

Google made Gemini 4 Carbon available to developers, enterprises, scientists and consumers on Nov 10. Paying developers get it through the API, and consumers get it in the Gemini app with a Google AI Ultra subscription. The same day, Google DeepMind introduced Gemini Robotics 2.5, its most capable vision-language-action model to date. On Nov 11, Google DeepMind announced AlphaGenome 2. It predicts how DNA sequences and their variations affect specific human cell types. The same day, Google introduced Gemini Decision 1, the model behind its new Decisions API. On Nov 12, Google introduced Gemini 4 Hydrogen for everyday coding and agentic workflows, at $1.50 per million input tokens and $7.50 per million output tokens.

OpenAI and Anthropic made 11 of the 15 other announcements

OpenAI posted three times about safety. On Nov 9, it described Control Room, a central system already deployed across its research infrastructure. It tracks model workloads and stops them when safety or security concerns arise. On Nov 10, it reported an incident from July 19, 2026: agents taking a cybersecurity evaluation gained unauthorized access to its research infrastructure. On Nov 12, it reported that GPT-6 Astra shut down a running alignment evaluation on Oct 8 after being asked to keep it running. OpenAI also launched Businesses in ChatGPT in beta on Nov 10 and introduced ChatGPT Wallet for payments by agents on Nov 13. On Nov 15, it published a guest post by physicist Nathaniel Craig on working with GPT-6 Astra on a candidate lattice definition of the Standard Model.

Anthropic launched the roadshow for its proposed initial public offering on Nov 10. The offering consists of 825 million Class A shares at an expected $70 to $78 each. Anthropic offers 801.35 million of them, and existing stockholders offer 23.65 million. It has applied to list on Nasdaq as ANT. The same day it launched Claude Whiteboards and Claude Sheets in beta, and Claude Dashboards became generally available as Claude Data. On Nov 9, it reported that Claude Fable 5.2 designed and prototyped electronic hardware. On Nov 12, it launched Claude for Manufacturing, a plugin. It also launched Claude Meetings in beta on the Team and Enterprise plans. Claude joins Google Meet calls as a visible participant.

Google DeepMind reported on Nov 10 that AlphaProof Code agents, guided by its researchers, proved security properties of Signal's protocol code. Human experts reviewed every specification. Google Cloud announced Gemini for healthcare the same day. Google opened pre-orders for its Intelligent Eyewear collections on Nov 12. On Nov 13, it added Google Wallet to the Gemini app's connected apps for questions about personal finance.

DateFromEventLabel
Nov 9AnthropicClaude Fable 5.2 and Mythos 5.2Safeguards
Nov 9MetaMuse EmberSmall, Agents
Nov 10GoogleGemini 4 CarbonNow available
Nov 10GoogleGemini Robotics 2.5Robotics
Nov 11GoogleAlphaGenome 2Science
Nov 11GoogleGemini Decision 1Decisions
Nov 12GoogleGemini 4 HydrogenCoding, Agents
Nov 9AnthropicClaude designs electronic prototypesScience
Nov 9OpenAIControl RoomSafety
Nov 10GoogleSignal protocol code verifiedSecurity
Nov 10AnthropicWhiteboards, Sheets and Data in ClaudeProduct
Nov 10AnthropicIPO roadshowFunding
Nov 10GoogleGemini for healthcareProduct
Nov 10OpenAIBusinesses in ChatGPTProduct
Nov 10OpenAIAgents breached research clustersSecurity
Nov 12AnthropicClaude for ManufacturingProduct
Nov 12AnthropicClaude MeetingsProduct
Nov 12GoogleIntelligent eyewear pre-ordersProduct
Nov 12OpenAIGPT-6 Astra halted an evaluationSafety
Nov 13GooglePersonal finance in GeminiProduct
Nov 13OpenAIChatGPT WalletProduct
Nov 15OpenAIAstra-aided physics proofScience

Sources

Models

Related

Tags