AI coding tools news
November 10, 2026 · 51 stories
The day's news for developers who use AI coding tools, picked and summarised from official changelogs, the media and Hacker News. Resets and limit changes are on the reset tracker.
Cursor adds multiplayer Projects and faster cloud agents
Cursor Projects now supports multiplayer collaboration. Cloud agents start three times faster with builds.
OpenAI Acquires Travel Startup Odessia
OpenAI has acquired Odessia, a travel startup created by Sonder founder Francis Davidson. OpenAI has not yet announced its future plans for the travel sector.
Google releases Gemini 4 Carbon model
Gemini 4 Carbon is now available to all users. Pricing is set at $2 per million input tokens and $10 per million output tokens, increasing to $4 and $20 respectively starting December 1st.
GitHub Copilot CLI v1.0.110-0 adds masked password prompts and credit top-ups
GitHub Copilot CLI version 1.0.110-0 introduces experimental masked dialogs for git, ssh, and sudo password prompts via the /experimental on command. It offers a session AI-credit limit top-up at 90% usage and displays recovery guidance when an unavailable account is selected in /user switch. The update also adds support for collapsing expanded reasoning by clicking anywhere, timeline visibility for compaction outcomes, and fixes handling of non-ASCII filenames in /diff.
Anthropic expands Claude to work directly inside enterprise applications
Anthropic is expanding Claude’s ability to work directly inside enterprise applications by allowing it to open full software interfaces and retrieve information.
All 7 Anthropic founders will each cash out about $250M in IPO
Each of the seven co-founders plans to sell the same number of shares in the IPO. This will bring each founder about $250 million in cash at the midpoint of the price range while the group retains company control.
Claude Code adds project organization and model configuration settings
Claude Code now allows users to organize tasks using projects in Claude Cowork. The update also enables users to modify model, effort, and thinking settings.
Claude Enterprise adds Security Posture data to Compliance API
The Compliance API now includes security posture data for Claude Enterprise organizations in beta. The Activity Feed in organization settings now records changes to attestations. A new GET /v1/compliance/organizations/{organization_id}/security_posture endpoint is available to retrieve current security results.
OpenAI introduces Businesses in ChatGPT
OpenAI introduces Businesses in ChatGPT to help users manage business information, view insights on business visibility, and convert interest into action.
Claude Code adds model configuration and updates thinking block handling
Claude Code adds model configuration options and updates the Messages API to handle thinking blocks with preserved thinking to protect against distillation. It also adds explanations for why Claude switched models during conversations with Opus 5 or Opus 5.5, alongside updates regarding the training data cutoff.
Google DeepMind Introduces AlphaProof Code
Google DeepMind has introduced AlphaProof Code. The model launch focuses on code generation and formal reasoning capabilities.
Kiro IDE 1.4.154 adds Approve for Me and Chat Permission Modes
Kiro IDE 1.4.154 introduces an 'Approve for me' feature that automatically handles routine actions while blocking risky ones. The Autopilot toggle is replaced by a permission menu offering Supervised, Accept file changes, and Approve for me modes. The update also includes organization-managed settings, Memory view sorting and filtering, and workflow run improvements.
Claude Code: Updates to Sheets, Whiteboards, and Motion features
Claude Code has updated its feature set, adding Claude Sheets (beta) and Claude Whiteboards (beta) while removing Claude Dashboards (beta). Claude Data now includes support for Sheets and Whiteboards, with some tiers also including Motion.
OpenAI launches verified Business Agents for ChatGPT
OpenAI is introducing verified Business Agents in ChatGPT to assist users with tasks like comparison shopping. The feature is currently available in beta for eligible e-commerce and online service businesses in the US.
Anthropic Launches Claude Whiteboards, Sheets, and Data
Anthropic introduced Claude Whiteboards, Claude Sheets, and Claude Data, allowing users to brainstorm, build sheets, and share data directly within the Claude interface. Additionally, partners can now embed apps directly into Claude to enable collaborative work on the same screen.
Anthropic launches Claude Meetings for workplace planning
Anthropic announced Claude Meetings, a feature that drafts discussion plans, maps timelines on a whiteboard, answers meeting chat questions, and accesses connected work apps for information. The capability is available as part of Claude’s functionality for workplace discussions.
GitHub Copilot CLI 1.0.109 adds CIDR support and Claude Fable 5.2
GitHub Copilot CLI version 1.0.109 introduces IPv4 and IPv6 CIDR support for sandbox host rules, adds Claude Fable 5.2 to the model picker, and improves subagent column readability on narrow terminals. The update also refines MCP server removal guidance, directory path handling, Windows path separators, and parallel file operation permissions.
GitHub Copilot CLI C++ language server now auto‑configures MSBuild builds
The Microsoft C++ Language Server for GitHub Copilot CLI now bundles its MSBuild compile commands extractor and invokes it automatically. An agent no longer needs to discover and run a separate step.
Claude Opus 5.5 tops OCR benchmark, Mistral OCR best value
We created eight scanned pages and evaluated them with 15 vision models, including Mistral OCR and Tesseract. Claude Opus 5.5 achieved the highest accuracy, while Mistral OCR offered the best value.
Anthropic launches its own AI-powered office suite
Anthropic announced the launch of its own AI-powered office suite.
GitHub Copilot pricing page adds Anthropic Claude Fable 5.2
The GitHub Copilot pricing and quota page now includes Anthropic Claude Fable 5.2. The entry appears multiple times on the updated page.
eBay blocks AI agents including Meta's Muse from browsing
eBay has blocked AI agents, including Meta's Muse, from browsing the platform. The change prevents agents from accessing eBay listings. No further details on timing or scope were provided.
Google adds two new specialized TPUs for confidential training and higher memory
Google announced two new tensor processing units (TPUs). They target confidential model training and increased memory capacity. The chips expand Google’s AI hardware portfolio for broader use cases.
SpaceXSI flips Grok API text watermarking to off by default
SpaceXSI changed the Grok API default setting to disable text watermarking. The change was made this past week. No other details were provided.
CNCF report: 75% of cloud‑native developers ship code with AI
The Cloud Native Computing Foundation report states that 75% of cloud‑native developers ship code with AI. The finding highlights widespread AI adoption among developers. No further numeric details are provided.
OpenAI and Upwork report AI’s effect on freelance work
New research from OpenAI and Upwork finds shifting demand for freelance work, new skill combinations, and larger budgets and more hours for AI-related projects.
Meta AI agent Muse now available on Apple Watch and Android Wear OS
Meta’s AI assistant Muse can now be used on Apple Watch. Users may interact via voice calls, text messages, or dictation. The feature is also available on Android Wear OS devices.
Claude Code update: preserved thinking change and project organization
The Messages API now changes how thinking blocks are handled to protect against distillation. Users can also organize tasks with projects in Claude Cowork.
Kora launches free Windows alternative for AI‑heavy users
Kora is set to launch as a free‑to‑use alternative to Windows. It targets users who rely heavily on AI models and agents.
Adobe announces major AI announcements at MAX 2026
Adobe announced major AI announcements at its MAX 2026 conference.
Spotify AI DJ adds new voices, personalities and chatter‑off option
Spotify Premium users can now customize the AI DJ with new voices that have different personalities. They can also turn off the commentary between songs.
Anthropic sets IPO price range of $70 to $78 for $1.404T valuation
Anthropic set an IPO price range of $70 to $78 per share. The company could reach an implied fully diluted valuation of $1.404 trillion if shares list at the top of the price range.
Adobe launches prototype builder tool in Firefly
Adobe is taking on Figma with a new prototype builder tool in Firefly.
Adobe adds automatic object removal to Photoshop, Lightroom, and Premiere
Adobe expanded its Creative Cloud apps and Firefly AI platform with new capabilities. The updates include automatic removal of objects, blemishes, and other unwanted distractions in Premiere, Photoshop, and Lightroom. No pricing or usage limit details were provided.
Adobe announces AI agents for design assistance at Max conference
Adobe announced that chatbots and AI agents can serve as design assistants and automate creative workflows. The announcements were made at Adobe's Max conference in Miami, Florida. The new features include allowing third‑party AI agents to access and modify user content.
Anthropic targets $1.4 trillion IPO
Anthropic confirmed its IPO could be one of the biggest ever. The IPO price is expected to be between $70.00 and $78.00 per share. The company aims for shares to start trading Thursday or Friday.
Google testing Agent Swarm mode for Antigravity
Google is developing Swarm for Antigravity, a bee‑marked command that coordinates multiple agents through a shared board and visible dashboard.
Otter AI adds support for 27 languages
Otter AI added support for 27 languages and announced an upcoming feature that can follow speakers when they switch languages.
Hugging Face exec warns cyberattacks could tighten open‑source AI security
Hugging Face executive Irene Solaiman warns that growing cyberattacks on open AI platforms could force tighter security restrictions, threatening the accessibility that makes open‑source communities valuable to developers worldwide.
Google Gemini adds memory summary with inspect, correct, delete features
Google introduced a Memory summary for Gemini that shows stored details. It includes tools to inspect, correct, and selectively erase entries. Users can view, edit, or delete their Gemini memory at will.
Da Mao Technology and Pudong Environmental Energy Launch Dawn Intelligent Computing Center Power Coordination Platform
Da Mao Technology and Pudong Environmental Energy announced a strategic cooperation and launched the Dawn Intelligent Computing Center power coordination platform. The platform aims to accelerate the evolution of eastern city‑type intelligent computing centers toward E‑AIDC.
TensorWorld raises over $100 M in funding within four months
TensorWorld物原科技 announced it secured over $100 million in financing within four months of founding. The funding round exceeded $100 million. The announcement was made in a news article on QbitAI.
Ant Group releases open-source LingBot‑Touch tactile foundation model
On November 10, Ant Group’s LingBot Technology announced the open‑source release of the LingBot‑Touch tactile foundation model. The model is made publicly available for developers to integrate tactile capabilities into AI applications. No pricing or usage limits were mentioned.
OpenRouter Limits and Pricing Page Updated
Google Cloud pricing page updated: $300 credit cannot cover Gemini services
The $300 credit cannot be used to pay for Gemini platform (formerly Gemini Enterprise Agent Platform, Vertex AI) services if you sign up for a free trial or upgrade to a paid account on or after November 16, 2026. The page was last updated on 2026-11-10 UTC.
GitHub Copilot CLI 1.0.109-1 adds CIDR support and Claude Fable 5.2 model
Added IPv4/IPv6 CIDR support for sandbox host allow and deny rules and added Claude Fable 5.2 to the model picker. Improved sandbox path settings to accept environment variables, use cache‑aware background wait timeouts by default, and fixed multiple UI and command‑line bugs.
Qianwen Office 2.0 launches Team Agent for multi‑person collaboration
Qianwen Office 2.0 was released with a Team Agent feature that supports multi‑person and agent collaboration. The update ends the previous single‑machine mode for office AI.
Seedream 5.5 speeds up and API price reduced
The new Seedream 5.5 model runs faster. Its API pricing has been lowered.
Claude Code v2.1.322 adds fast mode on Vertex
Added fast mode on Vertex in the global, US and EU regions, falling back to standard speed if Vertex refuses it.
Devin Desktop 3.12.43 changelog adds response handling and image size defaults
Version 3.12.43 was released. It adds a response that stops at the output limit before a tool call continues the turn and prevents GPT models from stopping to ask for input they can work out. Image reads now use an image_size default of 70 % of the longest side, at least 1024 px. Follow‑up prompts are no longer blocked when Devin Cloud usage is restricted.
Antigravity v2.30.0 adds Knowledge Work layout, voice mode and remote agents
Antigravity released version 2.30.0 with a Knowledge Work layout for research, writing and everyday tasks. Voice mode is now available to all users except business and enterprise accounts. Remote agents are rolled out to Google AI Ultra subscribers and to Google AI Pro paid‑plan subscribers. The new layout and features are not available on business and enterprise accounts.