AI News, Week of 9 November 2026: Gemini 4 Carbon, Claude Personal Finance and OpenAI Images 2.6 Lead 8 Launches

The short version: eight launches from 27 October to 9 November 2026 that change what a business pays for AI, or what it can hand off. Google announced Gemini 4 Carbon and the Genie 4 world model and cut the price of Gemini Omni video by 30%, OpenAI added GIF output to Images 2.6, Anthropic launched Personal Finance in Claude, new Enterprise security controls and more Claude Marketplace partners, and Meta gave its Muse app a voice. Every story links its primary source with the date it was published.

In this edition

1. Google DeepMind announces Genie 4, a world model for real-time interactive simulation with native audio (27 October 2026)
2. Google announces Gemini 4 Carbon, priced at $2 per million input tokens and $10 per million output tokens, with access limited to cyber defenders first (28 October 2026)
3. OpenAI introduces Images 2.6 in ChatGPT, Codex and the API, with GIF output and 50% faster generation (4 November 2026)
4. Anthropic adds security, developer infrastructure, creative and productivity partners to Claude Marketplace, which now lists over 25 agents and products (4 November 2026)
5. Anthropic launches Personal Finance in Claude, a read-only view of connected bank and card accounts, in US beta (5 November 2026)
6. Anthropic adds a security center, generally available inference hooks and new access controls to Claude Enterprise (5 November 2026)
7. Google releases Gemini Omni 1.2 Flash with 15-second video, multi-turn editing and a 30% lower price of $0.07 per second at 720p (5 November 2026)
8. Meta rolls out voice and expressive avatars in the Muse app on mobile, web and Mac (5 November 2026)

1. Google DeepMind announces Genie 4, a world model for real-time interactive simulation with native audio

Google DeepMind announced Genie 4, a world model that turns a prompt into a steerable, real-time simulation with generated audio.

• Genie 4 is a world model for real-time interactive simulation that generates native audio on the fly, including ambient sounds, music and interactive dialogue.
• Users can steer scenes with natural language, such as moving the camera, changing the weather or adding characters.
• Rollout has started to a limited group of trusted testers.
• Google says it is coming soon to Google AI Ultra subscribers in Project Genie and to early access partners in the Gemini API and AI Studio.

Vendor claims, not tested by us: Google describes the realism as unprecedented.

Why it matters: Genie 4 is not generally available yet, so teams planning around it should treat it as a preview and wait for Gemini API access before budgeting.

Source: Introducing Genie 4, Google DeepMind, published 27 October 2026.

Our review and score: Gemini on AI Review Rating.

2. Google announces Gemini 4 Carbon, priced at $2 per million input tokens and $10 per million output tokens, with access limited to cyber defenders first

Google called Gemini 4 Carbon its strongest model and priced it at $2 per million input tokens, but only cyber defenders can use it so far.

• Gemini 4 Carbon is described by Google as its strongest model, with cybersecurity and long-horizon capabilities.
• Google reports 88.2% on SEC-bench Pro, 54.8% on ExploitGym and 71.0% on CWE-bench.
• Launch pricing matches Argon at $2 per million input tokens and $10 per million output tokens, with cached input tokens priced at 95% off the input price.
• Access starts with a subset of cyber defenders and widens gradually to trusted partners.
• Google says it is taking part in the US government’s voluntary process for pre-release model testing.

Vendor claims, not tested by us: Google calls Carbon its strongest model and reports the benchmark scores from its own testing.

Why it matters: Developers cannot call Carbon yet, but the published price gives a ceiling to compare against when planning agent and security workloads.

Source: Introducing Gemini 4 Carbon, Google, published 28 October 2026.

Our review and score: Gemini on AI Review Rating.

3. OpenAI introduces Images 2.6 in ChatGPT, Codex and the API, with GIF output and 50% faster generation

OpenAI shipped Images 2.6, which can now output looping GIFs and generates images 50% faster.

• Images 2.6 is available in ChatGPT, Codex and the API.
• It adds animated GIF output alongside still images.
• OpenAI says images generate 50% faster than before.
• The API offers two models, GPT-Image-2.6 Flare and GPT-Image-2.6 Sunburst.
• With quality set to low, Flare can generate an image in as little as six seconds; Sunburst targets higher quality and edit consistency at medium or high quality.
• OpenAI says people now create 4 billion images per week across ChatGPT, Codex and the API.

Vendor claims, not tested by us: OpenAI says Images 2.6 keeps characters, products and brand details more consistent across edits.

Why it matters: Marketers and creators who need looping animations can now get them from the same tool they use for stills, and the Flare and Sunburst split lets API users trade speed against detail.

Source: Introducing Images 2.6, OpenAI, published 4 November 2026.

Our review and score: ChatGPT on AI Review Rating.

4. Anthropic adds security, developer infrastructure, creative and productivity partners to Claude Marketplace, which now lists over 25 agents and products

Claude Marketplace added security, observability and deployment partners, and enterprises can pay for them from their committed Anthropic spend.

• Claude Marketplace now lists over 25 partner agents and products.
• Eligible enterprise customers can apply a portion of their committed Anthropic spend toward these partner solutions.
• New security additions include Chainguard Harness, Checkmarx One, Cyera Agent Guardian, Rubrik Code Guardian and Tanium Cloud.
• Development infrastructure additions include Datadog Agent Observability and Netlify Enterprise.

Why it matters: Enterprise buyers with an Anthropic commitment can route part of it to partner tools, which changes how procurement can budget for security and observability around Claude.

Source: New agents and products on Claude Marketplace, Anthropic, published 4 November 2026.

Our review and score: Claude on AI Review Rating.

5. Anthropic launches Personal Finance in Claude, a read-only view of connected bank and card accounts, in US beta

Claude can now read your connected bank and card accounts with Personal Finance, a read-only US beta for paid plans.

• Personal Finance lets users connect bank and card accounts and ask questions in plain language.
• Access is read-only, and it cannot move, manage or invest money.
• Rollout starts in beta on the Pro, Max and Ultra plans in the US.
• Claude sees the bank name, account name, type, last four digits, balances, credit limits and up to two years of transactions for the accounts a user connects.
• Anthropic says conversations in the Personal Finance hub are not used to train its models, and that it has no ads or sponsored recommendations.
• Anthropic says it is not designed to give investment advice and that Claude can make mistakes.

Vendor claims, not tested by us: Anthropic says Personal Finance gives straight answers with no ads or sponsored recommendations.

Why it matters: US subscribers on paid Claude plans get a consumer money-questions feature, but because answers can be wrong, the figures are worth checking against the bank.

Source: Introducing Personal Finance in Claude, Anthropic, published 5 November 2026.

Our review and score: Claude on AI Review Rating.

6. Anthropic adds a security center, generally available inference hooks and new access controls to Claude Enterprise

Claude Enterprise gets a security center, generally available inference hooks and tighter access controls.

• A new security center page in Organization settings compares an organization’s Claude settings with Anthropic’s recommended configuration.
• Prompt hooks and pre-tool-call hooks are now generally available and send prompts and tool calls to an organization’s data security tools for a check.
• Centralized management gives a company one admin console for company-wide settings.
• New access controls let admins sign members out through the Admin API, manage the IP allowlist themselves and block personal Claude accounts in Claude Desktop on managed computers.
• Anthropic states the security center is guidance, not a certification or a guarantee of compliance with any law or standard.

Why it matters: Security teams can now apply their existing data loss prevention policies to Claude prompts and tool calls and review settings in one place, which removes a common blocker for enterprise rollouts.

Source: More visibility and control over Claude Enterprise for security teams, Anthropic, published 5 November 2026.

Our review and score: Claude on AI Review Rating.

7. Google releases Gemini Omni 1.2 Flash with 15-second video, multi-turn editing and a 30% lower price of $0.07 per second at 720p

Gemini Omni 1.2 Flash makes 15-second video clips and costs 30% less than the previous version.

• Gemini Omni 1.2 Flash supports native video creation and editing up to 15 seconds, up from 10 seconds with Omni 1.1.
• Duration can be set in one-second steps from 3 to 15 seconds.
• Google says the price is 30% lower than Omni 1.1, at $0.07 per second for 720p video.
• Clips can be refined through multi-turn conversational edits instead of regenerating the whole clip.
• The model is generally available on the Gemini API and Google Cloud, and is rolling out in Google Vids and Gemini for business.
• Drafts can be made at 360p or 720p and upscaled to 1080p or 4K.

Vendor claims, not tested by us: Google says Omni 1.2 Flash improves reference adherence and consistency across shots.

Why it matters: At $0.07 per second, a 15-second 720p clip costs about $1.05 at list price, which makes drafting and revising video by conversation cheaper than regenerating from scratch.

Source: Gemini Omni 1.2 Flash: 15-second video and lower pricing, Google, published 5 November 2026.

Our review and score: Gemini on AI Review Rating.

8. Meta rolls out voice and expressive avatars in the Muse app on mobile, web and Mac

Meta added voice and expressive avatars to the Muse app on mobile, web and Mac.

• Voice and expressive avatars are rolling out in the Muse app on mobile, web and Mac.
• Users start a conversation by tapping the phone icon in the top-right corner.
• Meta says there are hundreds of voices to choose from.
• The avatar can be shown voice only, alongside the conversation, or full screen.
• Meta says Muse works on tasks in the background while the user talks.

Vendor claims, not tested by us: Meta says conversations feel fluid and in real time.

Why it matters: Muse now competes on spoken conversation as well as text, which matters for users who want a hands-free assistant on desktop and phone.

Source: Give Your Muse a Voice, Meta, published 5 November 2026.

Our review and score: Meta AI on AI Review Rating.

How we picked and checked these stories

We lead with launches and tools a business can use or price, and leave out stories that only describe risk unless they change what a buyer should do. Creator posts and videos are where we hear about a story; they are never the source. Every fact above comes from the company’s own page, the GitHub API or a primary news outlet, read on 8 November 2026, and every story is dated inside 27 October to 9 November 2026. Numbers we could not test ourselves are labelled as vendor claims. Stories we could not verify on a primary source in that window were cut, and OpenAI’s dots launch is left out because the previous edition covered it. More editions: all AI News editions.

Your turn

Would you rather pay for AI from the cloud or ERP credits you already hold, or from a new contract of its own? Reply to our newsletter and tell us; the answers shape what we test next.

Get the next edition by email: join our free newsletter with the new AI tools we score.

Related reading

• AI News, week of 2 November 2026: the previous edition
• Our weekly AI tools digest for 19 October 2026: the tools we scored
• Open-source alternatives to paid AI tools

Sources

1. Introducing Genie 4 (Google, 27 October 2026), read 8 November 2026.
2. Introducing Gemini 4 Carbon (Google, 28 October 2026), read 8 November 2026.
3. Introducing Images 2.6 (OpenAI, 4 November 2026), read 8 November 2026.
4. New agents and products on Claude Marketplace (Anthropic, 4 November 2026), read 8 November 2026.
5. Introducing Personal Finance in Claude (Anthropic, 5 November 2026), read 8 November 2026.
6. More visibility and control over Claude Enterprise for security teams (Anthropic, 5 November 2026), read 8 November 2026.
7. Gemini Omni 1.2 Flash: 15-second video and lower pricing (Google, 5 November 2026), read 8 November 2026.
8. Give Your Muse a Voice (Meta, 5 November 2026), read 8 November 2026.