LLM/AI Changelog

Upload Larger Files in Copilot Studio Agent Builder
[Android, Windows, iOS, Web]
Agent Builder now supports file uploads up to 512 MB when creating agents, ideal for larger files.
Roadmap ID:
500375
Details:
What Changed:
This increases the upload size limit in Agent Builder to 512 MB, enabling use of larger files as grounding data.
Why:
Users requested more flexibility for grounding agents. Larger files reduce the need to split or compress documents.
Try This:
Drag and drop large documents such as training manuals into your agent project.
Create the agent and ask it to summarize information from uploaded files.
Why this matters:
Business Impact:
Allows enterprises to build agents with richer, domain-specific knowledge.
Personal Impact:
Complete your work without the need to split files or compress data.
Additional resources:
Learn:
Embedded file content

Use .NET client libraries to integrate Microsoft 365 Copilot APIs
[Web]
Offers official .NET SDKs for integrating with Microsoft 365 Copilot APIs.
Details:
What changed:
Released a .NET SDK for Copilot API calls, making integration easier.
Why:
Meets enterprise developers’ need for secure .NET integration options.
Try This:
Install NuGet package
Authenticate and initiate Copilot prompt calls
Process responses in your .NET app
Why this matters:
Business Impact:
Improves developer productivity and project speed.
Personal Impact:
Gives .NET teams direct hooks into Copilot capabilities.
Additional resources:
Learn:
Microsoft 365 Copilot APIs client libraries

Use Python client libraries to integrate Microsoft 365 Copilot APIs
[Web]
Adds official Python libraries for secure integration with Microsoft 365 Copilot APIs.
Details:
What changed:
Developers can now leverage Microsoft 365 Copilot APIs using Python SDKs.
Why:
Opens opportunities for AI workflows in Python-based applications.
Try This:
Install Python SDK
Authenticate using OAuth
Send prompts and parse results
Why this matters:
Business Impact:
Expands Copilot customization for enterprise developers.
Personal Impact:
Python devs gain first-class access to Copilot APIs.
Additional resources:
Learn:
Microsoft 365 Copilot APIs client libraries

Use TypeScript client libraries to integrate Microsoft 365 Copilot APIs
[Web]
Build custom integrations with Microsoft 365 Copilot APIs using official TypeScript SDK libraries.
Roadmap ID:
501574
Details:
What changed:
Official TypeScript SDK released for Copilot API integration in web apps.
Why:
Standardizes developer integration with Microsoft 365 Copilot features.
Try This:
Install SDK via npm
Authenticate using Microsoft Identity
Test API calls for Copilot prompts
Why this matters:
Business Impact:
Enables custom enterprise apps with AI-powered features.
Personal Impact:
Developers have a secure, documented method for advanced builds.
Additional resources:
Learn:
Microsoft 365 Copilot APIs client libraries

View and Manage SharePoint Agents with Greater Control
[Web]
Admins can now see all SharePoint agents in a single unified view and manage them more intuitively with a streamlined delete action.
Details:
What Changed:
The enhanced Copilot Control System UI introduces a centralized SharePoint agent inventory and replaces the previous block/unblock workflow with a clear, simplified delete option.
Why:
Admins requested more transparent oversight of deployed agents, and this update delivers easier tracking, review, and lifecycle management.
Try This:
Go to Agents and Connectors in the Copilot Control System.
Review all existing SharePoint agents and delete those that are outdated, unused, or no longer compliant.
Why this matters:
Business Impact:
Strengthens governance and operational hygiene by reducing risk from stale or misconfigured agents.
Personal Impact:
Saves time with straightforward, intuitive controls instead of complex, multi-step administration.

Build Copilot agents using organizational People data
[Windows, Web]
Developers can build agents in Copilot Studio that deliver personalized, context-aware responses using organizational People data.
Details:
What changed:
Added the ability for agent builders to access and use People data from your directory in responses.
Why:
Makes interactions more contextual and human-centric for better relevance.
Try This:
Enable People data in Agent Builder
Create an agent to answer org-specific directory questions
Test in Microsoft Teams
Why this matters:
Business Impact:
Improves accuracy and personalization in enterprise workflows.
Personal Impact:
Users get tailored answers courtesy of context-rich data.
Additional resources:
Learn:
Add knowledge sources to your declarative agent in Microsoft 365 Copilot

Use your organization’s approved assets in Copilot presentations
[Mac, Windows, Web]
Create branded PowerPoint slides by pulling images and templates from your company’s SharePoint asset library or Templafy integration.
Roadmap ID:
496366
Details:
What changed:
PowerPoint Copilot integrates with SharePoint Organization Asset Library and Templafy for approved, compliant visuals.
Why:
Ensures quality designs aligned with corporate branding guidelines.
Try This:
Configure SharePoint OAL or Templafy in Microsoft 365
Ask Copilot: “Create a marketing update deck using brand imagery.”
Why this matters:
Business impact:
Maintains brand identity across all content.
Personal Impact:
Saves design time by eliminating manual asset searching.
Additional resources:
Learn:
Create an organization assets library

Referenced sources cited in drafted content
[Windows]
Draft content with confidence as Copilot automatically includes citations, referencing the information in your text. This feature ensures accuracy and credibility, saving time on manual citation.
Roadmap ID:
380842
Details:
What changed:
Copilot in Word now automatically generates citations when incorporating external information into your drafts. This process highlights references directly within the text, eliminating the need for manual citation entry.
Why:
Proper attribution is vital for maintaining credibility and avoiding plagiarism. This enhancement simplifies the citation process, ensuring academic and professional integrity without added effort.
Try This:
While drafting in Word, ask Copilot: "Include citations for referenced content."
Use Copilot commands to review the generated citations for completeness and accuracy.
Compare the citation style used with your preferred or required academic formatting.
Why this matters:
Business Impact:
Enhances the reliability of professional documents by ensuring proper attribution without a complex citation process.
Personal Impact:
Save time creating references, allowing more focus on content quality and creativity.

Use .NET client libraries to integrate Microsoft 365 Copilot APIs
[Web]
Offers official .NET SDKs for integrating with Microsoft 365 Copilot APIs.
Details:
What changed:
Released a .NET SDK for Copilot API calls, making integration easier.
Why:
Meets enterprise developers' need for secure .NET integration options.
Try This:
Install NuGet package
Authenticate and initiate Copilot prompt calls
Process responses in your .NET app
Why this matters:
Business Impact:
Improves developer productivity and project speed.
Personal Impact:
Gives .NET teams direct hooks into Copilot capabilities.
Additional resources:
Learn:
Microsoft 365 Copilot APIs client libraries

Use your organization's approved assets in Copilot presentations
[Mac, Windows, Web]
Create branded PowerPoint slides by pulling images and templates from your company's SharePoint asset library or Templafy integration.
Roadmap ID:
496366
Details:
What changed:
PowerPoint Copilot integrates with SharePoint Organization Asset Library and Templafy for approved, compliant visuals.
Why:
Ensures quality designs aligned with corporate branding guidelines.
Try This:
Configure SharePoint OAL or Templafy in Microsoft 365
Ask Copilot: "Create a marketing update deck using brand imagery."
Why this matters:
Business impact:
Maintains brand identity across all content.
Personal Impact:
Saves design time by eliminating manual asset searching.
Additional resources:
Learn:
Create an organization assets library

Streamlined Copilot chat navigation with expanded history
[Windows, Web]
A redesigned navigation pane provides a cleaner layout and expanded chat history, making it easier to switch between conversations. This allows you to switch between conversations and pick up where you left off, perfect for fast-paced projects and multitasking.
Roadmap ID:
516570
Details:
What Changed:
The Copilot navigation pane highlights key agents, removes clutter, and expands visible chat history beyond the previous limit of five recent chats. Chat search is also more prominent for faster access to specific threads. This reduces interface clutter.
Why:
Users wanted an interface that feels organized and speeds up workflow. This redesign lets you quickly return to prior conversations without digging through menus.
Try This:
Open the navigation pane in Copilot to see the improved design.
Search for a previous project conversation using the updated chat search bar.
Pin your most important chats to keep them accessible during critical work sessions.
Why this matters:
Business Impact:
Standard, simplified navigation UI to help reduce downtime from searching old threads.
Personal Impact:
Enjoy a simpler, less cluttered workspace with easier access to past chats.

Today we’re rolling outYour Year with ChatGPT, an optional, personalized end-of-year experience that reflects on how you interacted with ChatGPT in 2025. It highlights high-level themes from your conversations and includes summary statistics about your usage over the year.
This experience is rolling out gradually throughout the day, so it may not be available to everyone immediately. It’s available to Free, Plus, and Pro users in United States, United Kingdom, Canada, Australia, and New Zealand as well as Go, Plus, and Pro users in India, and is not available on Business, Enterprise, or Edu plans.
To seeYour Year with ChatGPT, Memory and Reference Chat History must be turned on, and you must meet a minimum activity threshold. If you have very limited activity, you’ll only see basic chat statistics.
At launch, this experience is available in English in the United States, United Kingdom, Canada, Australia, and New Zealand.

Today we’re rolling outYour Year with ChatGPT, an optional, personalized end-of-year experience that reflects on how you interacted with ChatGPT in 2025. It highlights high-level themes from your conversations and includes summary statistics about your usage over the year.This experience is rolling out gradually throughout the day, so it may not be available to everyone immediately. It’s available to Free, Plus, and Pro users in the United States, the United Kingdom, Canada, Australia, and New Zealand as well as Go, Plus, and Pro users in India, and is not available on Business, Enterprise, or Edu plans.To seeYour Year with ChatGPT, Memory and Reference Chat History must be turned on, and you must meet a minimum activity threshold. If you have very limited activity, you’ll only see basic chat statistics.At launch, this experience is available in English in the United States, United Kingdom, Canada, Australia, and New Zealand.

In addition to selecting a base style and tone preset—like candid and friendly—you can now adjust specific characteristics individually.  From the ‘personalization’ pane in settings, you can adjust ChatGPT to:
Be more or less warmBe more or less enthusiasticUse more or fewer headers & listsUse more or fewer emojis

You can now browse and add approved apps in the newapp directory. Apps let you work with your tools and data directly inside a conversation—from interactive in-chat experiences to secure connections that allow ChatGPT to search and reference information from third-party services.As part of this update,connectors now appear in the directory as apps, making it easier to manage all your tools in one place. References across the Help Center have been updated to reflect this new terminology.Apps are available to all logged-in ChatGPT users, with availability and functionality varying by plan and region. Some apps or capabilities may not be available in certain regions or may require specific plan tiers.Additionally,developers can submit apps for review and publicationin the app directory, making approved apps available to a broader set of ChatGPT users.

In October, we introduced skills—a way to teach Claude repeatable workflows tailored to how you work. Skills are now easier to deploy, discover, and build withorganization-wide management for Team and Enterprise plans, adirectoryof partner-built skills, and an open standard (Agent Skills) so skills work across AI platforms.

Claude in Chrome is now available in beta to all paid plan subscribers, including Pro, Team, and Enterprise plans.
Claude Code integration:Build in your terminal with Claude Code, then test and verify in the browser with the Chrome extension. Claude can read console errors, network requests, and DOM state to help debug issues directly.Control browser actions from Claude Desktop:Start a task in Claude Desktop and let it handle work in the browser without switching windows.Record a workflow:Teach Claude your workflow by recording the steps, and Claude learns to repeat them. Useful for repetitive browser tasks you want to delegate.Console logs:Claude can now read browser console output, helping developers identify and fix errors without switching contexts.Admin controls for Team and Enterprise:Admins can now manage Claude in Chrome at the organization level:Enable or disable the extension org-wideConfigure allowlists and blocklists to control which sites Claude can accessManage Claude in Chrome in Admin settings > Browser extension

We’ve updated the Model Spec, our living document outlining intended model behavior, to strengthen and more clearly codify the principles that reflect how we build experiences for teen users.
New section: Under-18 (U18) Principles
ChatGPT’s new Under-18 (U18) Principles builds on the existing safety rules that apply to all users, adding age-appropriate guidance where appropriate for the developmental needs of teens, aged 13-17. This update clarifies how those rules are intended to apply in teen conversations, recognizing that teens benefit from clearer boundaries, reduced exposure to potentially harmful content and stronger real-world support when risks arise. The assistant should meet teens where they are, engaging with them in a respectful and transparent manner, while refusing to participate in self-harm, sexualized or violent immersive roleplay, dangerous activities, substance misuse or any efforts to conceal harm. When credible risks arise, the model should prioritize prevention and early interventions, offer safer alternatives and encourage involvement of parents, guardians and other trusted adults or professionals – making clear that AI can provide guidance and information, but cannot replace real-world care.
Other updates
This release also includes minor edits and clarifications for consistency and readability throughout the document.
More information can be found inthisblog post, and the latest version of the Model Spec is available atmodel-spec.openai.com.

- Launched Gemini 3 Flash Preview, `gemini-3-flash-preview`, delivering fast
frontier-class performance that rivals larger models at a fraction of the
cost. With upgraded visual and spatial reasoning, and agentic coding
capabilities. Read the documentation on some new features, including:

- [Multimodal function responses](https://ai.google.dev/gemini-api/docs/function-calling#multimodal)
- [Code execution with images](https://ai.google.dev/gemini-api/docs/code-execution#images)

Tasks let you set up automated prompts that ChatGPT can run for you.You can now manage all your tasks directly in Pulse:
See all tasks in one place in Pulse.Create tasks in chat or in Pulse—whatever’s easier in the moment.
Edit and review tasks in Pulse anytime after you create them.Please note that Pulse is only available in ChatGPT Pro.

* Launched Gemini 3 Flash Preview, `gemini-3-flash-preview`, delivering fast
frontier-class performance that rivals larger models at a fraction of the
cost. With upgraded visual and spatial reasoning, and agentic coding
capabilities. Read the documentation on some new features, including:

* [Multimodal function responses](/docs/function-calling#multimodal)
* [Code execution with images](/docs/code-execution#images)

Introducing Gemini 3 Flash, a major upgrade to your everyday AIWhat:Gemini 3 Flash is now the new default model in the Gemini app. It offers next-generation intelligence at lightning speed and represents a major capability upgrade over Gemini 2.5 Flash. This model provides PhD-level reasoning comparable to larger models and delivers a significant leap in multimodal understanding, allowing you to ask questions using images, audio, text and more.Gemini 3 Flash can be accessed in the Gemini app by selecting 'Fast' for quick answers and 'Thinking' to help solve more complex problems faster. Gemini 3 Pro is also available as 'Pro' in the model picker and remains the best choice for advanced maths and coding.Why:We want to make Gemini the most helpful AI assistant for everyone. By upgrading our default model to Gemini 3 Flash, we're bringing next-generation intelligence to all your tasks, helping you tackle your most ambitious projects, faster.

Tasks let you set up automated prompts that ChatGPT can run for you.You can now manage all your tasks directly in Pulse:
See all tasks in one place in Pulse.Create tasks in chat or in Pulse—whatever’s easier in the moment.
Edit and review tasks in Pulse anytime after you create them. Please note that Pulse is only available in ChatGPT Pro.

We’ve introduced a new and improved version ofChatGPT Images, powered by our best image generation model yet. With ChatGPT Images, precise instruction following and editing makes AI image creation not just more fun, but meaningfully more useful—from everyday edits to creative transformations.
All images you create with ChatGPT are automatically saved underMy imagesathttps://chatgpt.com/images, giving you a single place to browse, revisit, and reuse your work without having to sift through past conversations.
Learn more about ChatGPT Images in ourChatGPT Images FAQ.
Available onFree, Go, Plus, Edu, and Proplans. Business and Enterprise are coming soon!

We’ve introduced a new and improved version ofChatGPT Images, powered by our best image generation model yet. With ChatGPT Images, precise instruction following and editing makes AI image creation not just more fun, but meaningfully more useful—from everyday edits to creative transformations.All images you create with ChatGPT are automatically saved underMy imagesathttps://chatgpt.com/images, giving you a single place to browse, revisit, and reuse your work without having to sift through past conversations.Learn more about ChatGPT Images in ourChatGPT Images FAQ.Available onFree, Go, Plus, Edu, and Proplans. Business and Enterprise are coming soon!

- Released `gemini-2.5-flash-native-audio-preview-12-2025`, a new native audio model for the Live API. This update improves the model's ability to handle complex workflows. To learn more, see the [Live API guide](https://ai.google.dev/gemini-api/docs/live-guide) and [Gemini 2.5 Flash Native Audio](https://ai.google.dev/gemini-api/docs/models#gemini-2.5-flash-live).

* Released `gemini-2.5-flash-native-audio-preview-12-2025`, a new native audio
model for the Live API. This update improves the model's ability to handle
complex workflows. To learn more, see the [Live API
guide](/docs/live-api/capabilities) and [Gemini 2.5 Flash Native
Audio](/docs/models/gemini-2.5-flash-native-audio-preview-12-2025).

- Released `gemini-2.5-flash-native-audio-preview-12-2025`, a new native audio model for the Live API. This update improves the model's ability to handle complex workflows. To learn more, see the [Live API guide](https://ai.google.dev/gemini-api/docs/live-guide) and [Gemini 2.5 Flash Native Audio](https://ai.google.dev/gemini-api/docs/models/gemini-2.5-flash-live).

- Released `gemini-2.5-flash-native-audio-preview-12-2025`, a new native audio model for the Live API. This update improves the model's ability to handle complex workflows. To learn more, see the [Live API guide](https://ai.google.dev/gemini-api/docs/live-guide) and [Gemini 2.5 Flash Native Audio](https://ai.google.dev/gemini-api/docs/models/gemini-2.5-flash-native-audio-preview-12-2025).

- Launched the Interactions API in Beta. This API provides a unified interface
for interacting with Gemini models and agents. To learn more, see the
[Interactions API](https://ai.google.dev/gemini-api/docs/interactions) guide.

- Launched the Gemini Deep Research Agent in preview. It can
autonomously plan, execute, and synthesize results for multi-step research
tasks. See the [Deep Research](https://ai.google.dev/gemini-api/docs/deep-research) guide for
details.

We’re retiring the Voice experience in the ChatGPT macOS app on January 15, 2026. This change allows us to focus on more unified and improved voice experiences across our apps.
Voice will continue to be available onchatgpt.com, iOS, Android, and Windows app.No other ChatGPT features on macOS are affected.

Today, we’re releasingGPT-5.2, the next upgrade to the GPT-5 series. GPT-5.2 is smarter and more useful for both work and learning—while remaining just as enjoyable to talk to. As with GPT-5.1, we’re continuing to improve our main models for everyone on a regular basis to make them more useful and enjoyable to use.GPT-5.2 Instantis a fast yet powerful workhorse for everyday work and learning, with clear improvements in info-seeking questions, how-tos and walk-throughs, technical writing, and translation, all while retaining the warmer, more conversational tone introduced in GPT-5.1 Instant. Early testers noted its more clearly structured explanations where important info is highlighted upfront. It’s also more effective at supporting studying and skill-building, as well as offering clearer job and career guidance.GPT-5.2 Thinkingsolves harder work tasks more effectively and with more polish—particularly in spreadsheet formatting and financial modeling, alongside improvements in slideshow creation. Early testing showed gains in coding, summarizing long documents, answering questions about uploaded files, walking through complex math and logic step by step, and helping with planning and decision-making with clearer structure and thoughtful detail.GPT-5.2 Prois our smartest and most trustworthy model yet for difficult questions where a higher-quality answer is worth the wait. In early testing, it shows fewer major errors and stronger performance across complex domains like programming.All three models (Instant, Thinking, and Pro) have a new knowledge cutoff of August 2025. For users, this means GPT-5.2 starts with a more current understanding of the world, so answers are more accurate and useful, with more relevant examples and context, even before turning to web search.Read ourblogfor more details on examples and evals.Update for free and Go users:We’re removing automatic model switching for reasoning in ChatGPT. Previously, some questions were automatically routed to

* Launched the Interactions API. This API provides a unified interface for
interacting with Gemini models and agents. To learn more, see the
[Interactions API](/docs/interactions-overview) guide.

* Launched the Gemini Deep Research agent in preview. It can autonomously
plan, execute, and synthesize results for multi-step research tasks. See the
[Deep Research](/docs/deep-research) guide for details.

- Launched the Interactions API. This API provides a unified interface
for interacting with Gemini models and agents. To learn more, see the
[Interactions API](https://ai.google.dev/gemini-api/docs/interactions-overview) guide.

- Launched the Gemini Deep Research Agent in preview. It can
autonomously plan, execute, and synthesize results for multi-step research
tasks. See the [Deep Research](https://ai.google.dev/gemini-api/docs/deep-research) guide for
details.

- Launched the Interactions API. This API provides a unified interface
for interacting with Gemini models and agents. To learn more, see the
[Interactions API](https://ai.google.dev/gemini-api/docs/interactions-overview) guide.

- Launched the Gemini Deep Research agent in preview. It can
autonomously plan, execute, and synthesize results for multi-step research
tasks. See the [Deep Research](https://ai.google.dev/gemini-api/docs/deep-research) guide for
details.

Today, we’re releasingGPT-5.2, the next upgrade to the GPT-5 series. GPT-5.2 is smarter and more useful for both work and learning—while remaining just as enjoyable to talk to. As with GPT-5.1, we’re continuing to improve our main models for everyone on a regular basis to make them more useful and enjoyable to use.GPT-5.2 Instantis a fast yet powerful workhorse for everyday work and learning, with clear improvements in info-seeking questions, how-tos and walk-throughs, technical writing, and translation, all while retaining the warmer, more conversational tone introduced in GPT-5.1 Instant. Early testers noted its more clearly structured explanations where important info is highlighted upfront. It’s also more effective at supporting studying and skill-building, as well as offering clearer job and career guidance.GPT-5.2 Thinkingsolves harder work tasks more effectively and with more polish—particularly in spreadsheet formatting and financial modeling, alongside improvements in slideshow creation. Early testing showed gains in coding, summarizing long documents, answering questions about uploaded files, walking through complex math and logic step by step, and helping with planning and decision-making with clearer structure and thoughtful detail.GPT-5.2 Prois our smartest and most trustworthy model yet for difficult questions where a higher-quality answer is worth the wait. In early testing, it shows fewer major errors and stronger performance across complex domains like programming.All three models (Instant, Thinking, and Pro) have a new knowledge cutoff of August 2025. For users, this means GPT-5.2 starts with a more current understanding of the world, so answers are more accurate and useful, with more relevant examples and context, even before turning to web search.Read ourblogfor more details on examples and evals.Update for free and Go users:We’re removing automatic model switching for reasoning in ChatGPT. Previously, some questions were automatically routed to

Create polished videos faster with seamless editing and brand customization
[Web]
Turn text prompts or PowerPoint, Word, and PDF files into high-quality videos. Create professional videos quickly with easier editing, brand integration, and media customization.
Roadmap ID:
501560
Details:
What changed:
With the upgraded AI Video Creator in Microsoft 365 Copilot, use these new capabilities:
Transcript-based editing for greater control.
Access to your assets from OneDrive to update stock media.
Natural-sounding voiceovers for more authentic narration.
Brand color integration from Brand Kit.
A redesigned, scene-based structure for streamlined storytelling.
A cleaner editing experience for faster, intuitive workflows.
Why:
Video is a powerful medium, but creating professional-quality content is often time-consuming and technically challenging. This update removes roadblocks, so teams create compelling videos in a fraction of the time, using assets and brand elements they already have.
Try this:
In Microsoft 365 Copilot, select Create and upload a Word, PDF, or PowerPoint file to generate your first video draft.
Click on a sentence in the transcript to cut or move sections without timeline complexity.
Add your official colors from Brand Kit, and swap generic visuals with your OneDrive media for a branded look.
Why this matters:
Business Impact:
Reduce production bottlenecks and costs by empowering employees to create professional videos for training, marketing, or executive updates without third-party agencies.
Personal Impact:
Focus on the story, save hours on content creation, and eliminate the need for advanced video editing.
Additional resources:
Support:
Create a video with the Microsoft 365 Copilot app

Get AI-powered summaries with search views
AI Views creates AI-generated summaries to help you identify relevant content, quickly understand search results, and spend less time digging through documents.
Details:
What changed:
AI Views is now available in the Microsoft 365 Copilot app search experience. It uses AI and Microsoft Graph signals to generate key point summaries for your search results directly on the Copilot Search page.
Why:
Users often waste time viewing multiple documents to find the right information. Now users see an AI and Microsoft Graph-powered overview of search results inside the Search results page.
Try this:
Open the Microsoft 365 Copilot app and run a search for a project or topic you’re working on.
Hover over any search result to see the Overview button
View a summary containing context and use the input box at the bottom to Ask Copilot.
Why this matters:
Business Impact:
Teams makes decisions faster with quick access to context-rich summaries, improving productivity.
Personal Impact:
Get key information immediately, letting you focus on delivering results instead of hunting for data.

Access custom engine agents in multiple apps
[Web]
Use custom engine agents in Word and Excel for a unified Copilot experience everywhere you work.
Roadmap ID:
481136
Details:
What changed:
You can see more available custom engine agents across key Microsoft 365 surfaces.
Why:
Power users and developers can experience reduced fragmentation and improved workflow consistency.
Try this:
In Outlook, activate a custom agent for summarizing incoming correspondence based on enterprise policies.
Why this matters:
Business Impact:
Extends the value of custom Copilot agents across the Microsoft 365 ecosystem.
Personal Impact:
Offers a seamless experience without switching apps or rebuilding context.

Edit adaptive cards inline for a faster build process
[Web]
Developers can now make quick inline edits on adaptive cards without leaving their design environment, reducing turnaround times.
Details:
What changed:
Action.Execute now supports inline editing for adaptive cards.
Why:
Custom apps allow for unblocked rapid iteration.
Try this:
While building a card, adjust field labels and actions inline instead of exporting and re-importing.
Why this matters:
Business Impact:
Speeds up partner development cycles for faster app rollouts.
Personal Impact:
Eliminates frustrating small text or layout changes.
Additional resources:
Learn:
Allow inline editing of Adaptive Card responses

Better security insights help you build confidently
[Web]
Makers can now see an agent’s detailed protection status and actionable recommendations in Copilot Studio.
Details:
What changed:
Enforce protections accurately with a richer visualization of security posture for agents.
Why:
Creators are empowered to manage compliance and proactively minimize risks.
Try this:
In Copilot Studio, check the security summary for your agent and apply suggested actions to meet policy.
Why this matters:
Business Impact:
Maintains compliance without slowing innovation.
Personal Impact:
Gives creators peace of mind knowing their agents meet enterprise security standards.
Additional resources:
Learn:
Agent runtime protection status

Add multiple SharePoint agents in one Teams conversation
[Web]
Include more than one SharePoint agent in a single Teams chat, meeting, or channel. Teams and SharePoint now work together better for smarter collaboration.
Roadmap ID:
481136
Details:
What changed:
Teams supports multiple SharePoint agents in a single group chat or channel.
Why:
Experience richer scenarios where multiple libraries or sites provide simultaneous insights.
Try this:
In your project channel, add two different SharePoint agents to get instant answers about separate libraries.
Why this matters:
Business Impact:
Keeps all content conversation in context, reducing silos across different sites.
Personal Impact:
Stay focused without the need to use multiple chats or apps to gather information.
Additional resources:
Support:
Share an agent from SharePoint in Teams

Multilingual support in Copilot for Viva Glint
[Web]
Copilot in Viva Glint now understands and responds in multiple languages, helping employees interact in their preferred language.
Roadmap ID:
508531
Details:
What changed:
Copilot for Viva Glint detects the language of a user’s input and responds in that same language.
Why:
Global teams often face challenges when a tool’s language support is limited. This update improves accessibility and usability, enabling employees to share feedback or gain insights without language barriers.
Try this:
Type a prompt in your preferred language within Viva Glint Copilot.
Review the response to confirm it matches the language you used.
Share this experience with multilingual teams to streamline feedback processes.
Why it Matters:
Business Impact:
Enhances inclusivity and compliance, making it easier to manage employee feedback across multiple regions.
Personal Impact:
Saves time and reduces friction for employees who can now interact with Copilot in the language they’re most comfortable with.