Hello, Moveworks Community,We’ve added a new field to the Moveworks Data API that makes the Bot Friendly Name configured in Moveworks Setup available through the API.Starting today, the /interactions API will capture this value in the platform_name field. This allows you to capture interactions happening on specific platforms.  To view the Bot Friendly Name configured for your chatbot, navigate to Moveworks Setup → Manage chat bots → Bot friendly name 

OverviewMoveworks is simplifying the Setup experience by removing configuration settings that are still displayed in Moveworks Setup but no longer affect how your instance behaves. The first set of these settings has now been removed. This cleanup is ongoing, and further updates will be shared as they roll out.What's ChangingThree settings that no longer control any functionality have been removed from Setup:Access Account — "Time waited before delivering link" and "Time waited before closing the ticket" Channel Resolvers → Channel Settings — "Enable skill in dry-run mode" (under Additional Controls)These settings were no longer connected to any active functionality, so their values had no effect regardless of how they were configured. You may notice they are no longer visible on these pages. What Stays the SameYour Moveworks experience is unchanged. Because these settings were not driving any behavior, their removal has no impact on how your assistant operates, and no existing configuration is affected. All other settings on these pages continue to work exactly as before. Required ActionsNo action is required. This cleanup is still in progress, and we'll share further updates as additional settings are reviewed and removed. If you have any questions or encounter any issues, please file a case through the Now Support portal.

We've upgraded the LLM Actions you already use, generate_text_action and generate_structured_value_action, with a boost in performance and a batch of powerful new capabilities. Best of all, it's a drop-in upgrade. Everything you've already built keeps working exactly as before, with nothing you need to change. ⚡ Better performance, no work requiredYour existing Compound Actions and Conversational Processes now benefit from improved model quality and efficiency automatically, especially on reasoning-heavy tasks. Same inputs, same outputs, better results.✨ What's new📄 File input Pass documents straight into the model, no preprocessing required. Supports PDFs, spreadsheets (Excel/CSV/TSV), rich documents (Word/RTF/ODT), presentations (PowerPoint/Keynote), and a wide range of text and code formats. 🎞️ Images and files together Combine images and files in a single request, and pass multiple of either as a list, from a single List[File] slot or mapped from separate File slots: file: [data.contract, data.addendum] 🌐 Web search Turn on enable_web_search to let the model pull live information from the web into its context instead of relying only on its training data. Want to keep it focused? Optionally scope results with web_search_allowed_domains or web_search_blocked_domains (up to 100 domains each). Available on all models except gpt-5 with minimal reasoning. 💡 TipFor image and file inputs, we recommend using gpt-5 or newer for the best results.🛠️ What you can build- OCR and document Q&A- Comparing multiple files (for example, contract vs. addendum)- Structured extraction from receipts, invoices, and forms- Answering questions that need up-to-date, real-world information🛡️ A note on safe useWeb search and file input let the model read content from outside sources. When designing flows that mix sensitive data, external content, and outbound actions, follow the security guidance in the docs to stay protected against prompt-injection risks.📖 Full details, examples, and the model reference: https://docs.moveworks.com/agent-studio/actions/llm-actions 

We've released the MLS Response Language Consistency Rule model upgrade to Standard customers.Non-English users sometimes saw responses that were mostly in their language but still contained stray English phrases — status updates ("Filing your ticket…"), confirmations, and live-agent handoff messages that the assistant had been instructed to repeat verbatim. We've updated the AI Assistant so these phrases are now produced in the user's own response language instead of being copied in English, for a more consistent multilingual experience.In offline evaluation on previously-affected conversations, the assistant went from leaking English in every case to correctly localizing ~90% of them, with no regressions. The improvement rolls out in a phased way across model-upgrade tiers. (A small set of fixed workflow status messages follows a separate localization path and isn't covered by this fix.)Read more about the release here.Learn more about our model upgrade program here.

🎉Coming Soon: BriefMe Multimodal Understanding for PDFs. 🎉We're excited to share that an important upgrade to BriefMe will be available early-mid August! What is BriefMe Multimodal Understanding for PDFs?BriefMe can now "see" and interpret the image elements inside uploaded PDFs. Instead of just reading the raw text, your assistant can now contextually understand these types of images - pictures, charts, tables, and complex diagrams. This means you can upload visually enriched files and have a seamless, more accurate conversation with the assistant about the document. Who it’s for: Initially, this feature will be available to BriefMe (Slack/Teams) users in US, CA, and EU regions. In the coming months, we plan to make it available in all other regions. Also, we plan to extend Multimodal Understanding to DOCs and PPTs in the coming weeks. What this unlocks:Process Visual Data: Effortlessly extract answers from trend lines, stock curves, and complex tabular images. BriefMe can perform cell value lookups and accurately interpret X/Y plots. Complex Diagram Interpretation: Process flowcharts, IT architecture diagrams, and org charts are now fully readable. BriefMe understands blocks, directional arrows (top-down, left-right), dotted lines, and color schemes. Cross-Image Analysis: Upload multiple visual PDF files at once—like a generic process flow PDF and a specific departmental org chart PDF — and ask questions that require connecting the dots between images across files. Multilingual Support for Images also: We currently support multilingual abilities for text. We are extending that ability for images as well. We will be able to analyze and summarize visual flows even if the embedded text is in a non-English language. (For example, uploading a Spanish chemical plant flowchart and asking for an English summary). A few ways you can use this feature:HCM & HR: Upload a company org chart to instantly analyze reporting structures, or upload a complex holiday calendar PDF to filter mandatory vs. optional dates. IT & Engineering: Upload an IT support tier diagram to find the right escalation path, trace data paths in architecture diagrams, or even upload a PDF document with screenshots of code to get rewriting suggestions. Finance: Upload a chart from an annual report and ask BriefMe to calculate specific investment growth over time based on the trend line. Marketing: Upload a slide deck in PDF format and extract key insights from the entire deck.  Learn more here. Read how to get started here. Questions? Drop them in this thread! Our team and the broader community will jump in. Once you start using Multimodal Understanding for PDFs, we'd love to hear from you! 

We’ve rolled out an improvement to the AI Assistant that ensures a smoother, more natural experience for your non-English speaking users. Previously, when users conversed with the assistant in their preferred language, they would occasionally see English phrases during system-guided moments—such as status updates (e.g., "Filing your ticket..."), task confirmations, or live-agent handoff messages. This happened because the assistant was instructed to repeat those specific phrases verbatim.We have updated the AI Assistant's logic so these system phrases are now dynamically translated into the user's active conversation language.The Benefits A Seamless Localized Experience: Users will no longer see abrupt, untranslated system messages. Higher Clarity & Trust: Key information in the chat are now fully readable and clear to non-English speakers. Proven Accuracy: In our evaluations of previously affected conversations, the assistant successfully localized approximately 90% of these system phrases. This update is fully automated and requires no actions from admins or users.Note: This update is rolling out in a phased approach. Refer to this article for timing. A very small subset of fixed workflow status messages follows a separate system path and is not covered by this specific update, but the vast majority of conversational phrases will now be fully localized.

The Enterprise Search App will get smarter about the sources of your answers. It will now recognize the kind of question you're asking and give extra weight to the systems most likely to have the answer — so the most authoritative source rises to the top. Ask about a customer's ARR and Salesforce results lead; ask for a teammate's latest message and Slack comes first; ask about a bug fix and the relevant GitHub PR is front and center. For broad questions that could live anywhere, Search will keep its existing ranking — applying this new source awareness only when it's confident it will help.What this means for you: Even more relevant results — answers from the most authoritative system rise to the top. Reach the right answer with less scrolling across results. Works across all your connected systems, with minimal impact on search speed.This update will be rolled out by July 31, and requires no action from admins or end users. You'll simply notice even sharper results. 

Moveworks Enterprise Search now offers the Max Capacity journey for connecting your content at scale, generally available to all organizations. You will find it in Moveworks Setup under Search → Configure Search → Max Capacity. What Max Capacity offers, across nine systems (Box, Confluence Cloud, Google Drive, Outlook, ServiceNow, SharePoint Online & OneDrive, Slack, Unily, and Zendesk):Up to 100 million documents per connector. Automatic connector validation during setup, so access requirements issues are caught early. A single guided setup journey that uses your existing identity and sets up permissions for you, so there is no extra configuration to setup. Permission ingestion on every indexed connector, so employees only see content they are allowed to see. Live search on SharePoint & OneDrive, Slack, and Outlook — results are fetched from the source in real time, with nothing stored in Moveworks. Private-content search on Google Drive and Box. An Audience configuration that sits on top of mirrored permissions to roll out content in phases to your employees.Important Callouts:Identity configuration: the Max Capacity journey sets up identity automatically only if your organization already has an identity configuration and its joining key resolves to email. If you don't have an identity configuration yet, or you identify users by a non-email attribute (for example, an employee ID) that isn't mapped to email, please set up the identity configuration for the new connector manually — see Configure User Identity. Ingestion schedule: ingestion begins about 12 hours after you save a configuration; please allow 18-24 hours before testing search results. Nomenclature: in the Connectors page these connectors may currently show a "(Next Gen)" label. We'll be renaming it to "Max Capacity" this week.For new setups on any of the nine systems Max Capacity supports, we recommend the Max Capacity journey going forward.Each connector's setup guide lists its specific capabilities — review the guide for the system you're connecting before you start.Learn more:Choosing Your Journey: https://help.moveworks.com/ai-assistant/search-connectors#choosing-your-journey Max Capacity Setup: https://help.moveworks.com/ai-assistant/max-capacity-setup Per-connector setup guides: SharePoint and OneDrive ServiceNow Google Drive Box Confluence Cloud Slack Outlook Unily Zendesk Request a connector or share feedback: https://community.moveworks.com/ideas

MCP Workspace is going into controlled availability!You can connect MCP servers and bring more of the tools your team already uses into the assistant. That opens up a lot of useful stuff to build.Want to be considered? Sign up here.

Today we updated the Standard tier to GPT-5.4.Read more about the GPT-5.4 upgrade to the Moveworks Reasoning Engine here. Read more about the model upgrade program here.

Hello Moveworks community,We've improved how ambient agent notifications are represented in the Data API.What's changingAmbient agents that deliver plain-text notifications will now be correctly categorized in the Data API /conversations entity as CONVERSATION_TYPE_AMBIENT_AGENT_NOTIFICATION, with use-case details (usecase_id, usecase_name) populated.Previously, these plain-text notifications were not classified under this type and instead appeared as CONVERSATION_TYPE_UNSPECIFIED_SYS_EVENT without use-case details — making it difficult to identify and report on them.What you'll noticeAn increase in CONVERSATION_TYPE_AMBIENT_AGENT_NOTIFICATION volume. A corresponding decrease in CONVERSATION_TYPE_UNSPECIFIED_SYS_EVENT for the same records. Populated use-case details for these notifications going forward.Things to keep in mindForward-looking only — this applies from the release date onward. Historical data cannot be backfilled. If you have filters built on CONVERSATION_TYPE_UNSPECIFIED_SYS_EVENT or that count CONVERSATION_TYPE_AMBIENT_AGENT_NOTIFICATION, you'll see a one-time shift on the release date (starting today) . This is the intended correction, not an anomaly. Plain-text notifications contain no interactive elements, so only "notifications sent" counts are available — there is no click/interaction tracking for these conversations.

Hello Moveworks community,  Any Setup configuration that reads from or writes to an external system needs to point at a registered connector. Until now, most configurations let you pick that connector from a dropdown of your registered connectors — but a handful still asked you to type the connector name in as free text.That inconsistency was more than cosmetic. Free-text entries weren't validated until runtime, so a small casing difference, a typo, or a connector that had since been renamed would pass silently at save time and then fail later — causing ingestion failures and, in some cases, bot outages that were hard to trace back to the source.What's changedWe've standardized connector selection on the dropdown pattern everywhere. The following four configurations now use the same validated connector dropdown as the rest of Setup:- User Identity → Advanced settings → Bot account- User Identity → Advanced settings → Manual user management- Triage → Model retraining- Triage → Metadata cacheYou now select from your registered connectors instead of typing a name.What this means for you- No more silent failures — the connector is validated when you save, not at runtime. Typos and casing mismatches are eliminated.- Rename-safe — because configurations reference the registered connector itself, renaming a connector no longer breaks these configs.- A consistent experience — connector selection now looks and behaves the same across every Setup surface.Do I need to do anything?No. Your existing values were automatically migrated to the new dropdown — there's no action required and no interruption to your current configurations. Next time you open one of these pages, you'll simply see the connector dropdown in place of the free-text field.New connector selection   

Hello Moveworks Admins,We’re excited to announce that Self-Serve Validation Failure Alerts will be available in Moveworks Setup on 6th July 2026. This release introduces enhanced validation alerts for Users and Forms resources, giving admins greater visibility into ingestion validation failures along with the ability to safely override certain validation failures directly from Setup when appropriate. ▶️ Validation OverrideTo help protect against Large Data Changes (LDCP), Moveworks performs implementation validation checks on ingestions to detect significant or unexpected changes in data. For e.g., if the number of users decreases by more than a threshold compared to the previous ingestion, a validation alert is generated and the ingestion cannot proceed until the issue is resolved or the change (if expected) is overridden by Moveworks FDEs.A validation override is used when a validation alert is triggered, but the detected change has been reviewed and confirmed to be expected. Overriding the alert allows the ingestion to proceed and establishes the latest changed data as the new baseline.Note: Only override a validation alert after confirming that the underlying data is accurate and that the detected change is intentional. A validation override is irreversible—the updated ingestion state becomes the new baseline and cannot be undone. If the alert indicates an unexpected or unintended change, the underlying issue should be investigated and resolved instead. 🌟 Key BenefitsIdentify validation failures with detailed alert descriptions and actionable guidance. Safely override supported validation failures, with mandatory justification for every override action. Access resolved alerts to review override details, including the user who performed the override. Explore common causes and recommended actions to quickly troubleshoot alerts. 🧭 What’s New?1️⃣ Enhanced Validation Alerts: Moveworks now provides detailed validation alerts for common ingestion issues across Users and Forms, including:User Identity Significant change in user records Significant drop in user attributes Significant change in user record IDs Duplicate user records Schema validation errors Zero users ingested Record uniqueness failures Forms Significant drop in forms Significant drop in form field options Each alert includes a detailed description to help you understand the issue and determine whether the change is expected or requires investigation.2️⃣ Self-Serve Override FunctionalityFor supported validation failures, admins can now override the alert directly from Moveworks Setup by providing a mandatory justification. Override is available for:Significant change in user records Significant drop in user attributes Significant change in user record IDs Significant drop in forms Significant drop in form field optionsNote:If the change is expected, selecting Override accepts the current ingestion as the new baseline and resumes processing. If the change is unexpected, select Troubleshoot to investigate and resolve the underlying issue instead. 3️⃣ Redesigned Alerts ExperienceThe Alerts page has been redesigned to make validation failures easier to understand and act upon.New Impact columnInformational – An issue was detected, but Moveworks continues serving users from the previous successful ingestion. These alerts do not require an override and automatically resolve once the underlying issue is fixed. Blocking – Ingestion has been paused because a validation rule detected a significant or unexpected change. Resolve the issue or override the alert (where supported) to continue.Additional UI improvements include:Improved the alert screen with a more intuitive and streamlined user interface. One-click access to troubleshooting guidance and recommended resolution steps. Integrated Override action for supported alerts Renamed Archived Alerts section to Resolved Alerts Moved the Acknowledge button within the alert details dialog ✋🏻 Important CalloutsValidation Failure Alerts are currently available only for Users and Forms resources. Override is available only for supported validation failure types. Validation failures such as duplicate users, schema validation errors, zero users ingested, and record uniqueness failures require troubleshooting before ingestion can continue. Override actions are permanent and cannot be undone. After a successful override, the alert is moved to the Resolved tab. The new data baseline is accepted after the next successful ingestion. Have questions or feedback? Please post them on our Community Forum, where our team will be happy to assist you. Thank you,Moveworks Setup Team

We've released the Improvements to Bold Text in Responses model upgrade to Standard customers.Web chat responses used to over-apply bold — emphasizing IDs, ticket numbers, dates, metrics, and whole phrases of prose. When everything is bold, nothing stands out, and screen readers don't announce bold at all, so emphasis was lost for users relying on them. We replaced the permissive bolding behavior with a strict allowlist that treats bold as a structural signpost (headings and short run-in labels) rather than an emphasis tool. In a 250-sample side-by-side evaluation on real production traffic, this reduced inappropriate bold by ~88% (from 5.15 to 0.61 spans per response) with zero regressions. The result is cleaner, more scannable, and more accessible responses in web chat.Read more about the release here.Learn more about our model upgrade program here.

We are excited to share that the Conversations API is soon launching into the Controlled Availability phase!For the first time, you can put the Moveworks Assistant behind your own surfaces, your own desktop and mobile apps, internal portals, orchestration layers, and test harnesses, over a simple REST interface, without having to own a native chat experience.Why we built itMoveworks has always powered the surfaces employees already use (Slack, Teams, embedded web chat, etc.). The Conversations API extends that reach: it is the programmatic interface to the Assistant, bringing knowledge lookup, plugin execution, multi-turn conversations, and proactive messaging to any surface you build. It is the same Assistant your employees know, now available wherever you need it.What you can do in Controlled AvailabilityMulti-turn conversations with server-managed context, so you never resend prior turns. Slot filling and disambiguation are handled for you. Assistant-driven search with citations grounded in your ingested knowledge. Agentic workflows: run Agent Studio plugins and native actions (ticketing, PTO, access provisioning) through conversational flow. Proactive delivery: approvals, ambient agent messages, and Employee Comms campaigns delivered as events. Richer button interactions: slot disambiguation, confirmation groups, single and multi static selects, URL and redirect buttons, and native plugin follow-up buttons. Feedback submission (helpful / unhelpful plus free text) attributed back to Insights. Standardized authorization: Authorization Code flow for core IDPs (Okta, Azure, etc.) and self-signed JWT for service and agent requests with user impersonation.The surface is built on four constructs, Conversations, Responses, Messages, and Events, supporting synchronous, asynchronous, and proactive interaction patterns.Who it is forControlled Availability is a great fit if you are looking to:Bring Moveworks to custom desktop, mobile, or portal surfaces beyond Slack and Teams, including branded front doors for your full employee base. Embed Moveworks inside an internal router or orchestration layer that decides which AI handles a query. Programmatically evaluate Assistant responses on a cadence.You can deploy your integration to a pilot cohort of end users to test your integration, and further to your broader organization!A Note on Controlled AvailabilityControlled Availability is intentionally less opinionated about UX at the API layer than our native Slack and Teams experiences. A few capabilities (in-modal form rendering, handoff fallback for legacy deflections, and per-integration analytics tagging) are not yet supported, and your feedback during this phase will directly shape what ships at GA.How to sign upSign-ups are open now. Access is curated by our Product team, and accepted nominees will begin onboarding on July 20. To request access, fill out the sign-up form here. We will review your submission and reach out to discuss fit and next steps.What to expect before you sign up: This is API access, not a turnkey feature. To get the most out of Controlled Availability, you should come in with a pilot integration in mind that you want to build, and have technical resources or developers available to build it on top of the API.We cannot wait to see what you build! Questions or ideas? Drop them in the comments below.

Installing a plugin used to be messier than it needed to be.If the target org already had a connector, HTTP action, compound action, or conversational action with the same name, Agent Studio created forced you to create another copy by renaming it. One install could leave you with Create Ticket, Create Ticket (1), Create Ticket (2), then a bunch of manual cleanup to figure out which one was the real thing.That changes today. Reuse Existing Assets on Plugin Install is now GA for all customers.When you install a plugin from Marketplace or import one from another environment, Agent Studio now checks for matching assets by name in the target org. For each match, you can choose:Incoming asset ChoiceServiceNow connector Use existing or configure newCreate ticket action Use existing or configure newNotify user action Use existing or configure newPick Use existing and the plugin points at the asset already in that org. Anything new still gets created as new.What this fixes:→ Less asset sprawl after installs→ Less manual rewiring after sandbox-to-prod promotion→ Same install flow for Marketplace installs and environment importsThe important caveat: this is reuse, not overwrite. Agent Studio does not modify the asset you reuse, and this does not add version history or rollback. If you need a separate copy, choose Configure new.You'll only see existing assets you're allowed to reuse.If plugin installs have left your org with a trail of duplicate actions, try the new flow and tell me where it still feels rough.

🚀 Big newsWe just made getting answers in EmployeeWorks even faster and easier to share. No more telling colleagues, "Go to EW and type exactly this."Starting today, EmployeeWorks now officially supports Prefill Deep Links!What’s New?You can now construct a custom hyperlink that opens EmployeeWorks with a query already provided to the Moveworks Assistant. When a teammate clicks it, they land right in their own EW portal with the query already being worked on by the Assistant.Here is the URL format to build your own:https://{your-instance}.service-now.com/aiux/employeeworks/chat?mw_route=%2Fassistant%3Fmessage%3D{your-query}Where {your-query} is your question, URL-encoded (spaces become %20, question marks become %3F).💡 See It In ActionCheck out how these look in practice: How do I create an email signature? 👉 https://{your-instance}.service-now.com/aiux/employeeworks/chat?mw_route=%2Fassistant%3Fmessage%3DHow%20do%20I%20create%20an%20email%20signature%3F When clicked, will redirect the user into Moveworks with the provided query. I got locked out of my laptop what do I do? 👉 https://{your-instance}.service-now.com/aiux/employeeworks/chat?mw_route=%2Fassistant%3Fmessage%3DI%20got%20locked%20out%20of%20my%20laptop%20what%20do%20I%20do%3F The response will return just like if they queried Moveworks directly. ✨ Bonus Feature: Want to deep-link to an Enterprise Search result instead of the assistant? Use the same pattern but swap the route to mw_route=%2Fsearch%3Fquery=...How to Use It To Supercharge Your Workflow "Try This in EW" Buttons: Embed prefilled links into your internal documentation, wikis, runbooks, or onboarding labs so users can test queries instantly. Share Perfect Prompts: Crafted the perfect question that yields the exact answer you need? Send it to a teammate so they can run it on their own account. Intranet & Banner Shortcuts: Link directly from your intranet tiles or Self-Service Portal (SSP) banners straight to a specific EW use case rather than tgeneric homepage. ⚡ Live Right NowPrefill deep links are officially live and deployed across all EmployeeWorks instances. Start building your links and streamlining your team's workflows today!

We're retiring the legacy Control Center (the old admin interface) on July 10, 2026.All administration and configuration for your Moveworks environment is now available in My Moveworks at my.moveworks.com, which leverages your company's SSO for access. My Moveworks is a full replacement for the legacy Control Center — there is no loss of functionality, and no migration work is required on your side.If you've been accessing the older admin experience, you've already been automatically redirected to the new experience over the past two months. After July 10, the legacy URL will no longer be available.What you need to do: Nothing in most cases. If your organization experiences any issues during this transition, please open a Moveworks Support case and our team will help! Old Experience Page being sunset. Currently, users are being redirected to the new experience.New ExperienceNew experience users have been automatically redirected to for the past 2 months.  

 We've made it faster to get to the exact interaction you're troubleshooting. Chat Playback now lets you look up history by Trace ID, in addition to Conversation ID.Why it helpsPreviously, you needed the full Conversation ID to replay an interaction. Now, if you're already in your logs investigating an issue, you can grab the ID right there and jump straight into Chat Playback — no extra steps to track down the conversation.How it works1. In Chat Playback, use the dropdown to pick your lookup type:  - Conversation ID — loads the entire conversation, so you can step through every turn.  - Trace ID — takes you straight to the single interaction turn.2. Copy the ID from either the logs view:  - Built-in Plugin Logs → the Trace ID column  - Agent Studio Logs → the UUID column3. Paste it into Chat Playback and press Enter.That's it — you'll land right on the conversation or the specific turn you're after.📖 Full details in the docs: AI Assistant Chat Playback: Link  

 We've shipped two improvements to Moveworks Setup that make configuring your assistant smoother and troubleshooting conversations dramatically faster.---1. A simpler, smarter Ticketing Configuration wizardSetting up a ticketing integration just got cleaner. The Ticketing Configuration wizard now uses a single, context-aware action button in place of the separate Save and Next buttons.Instead of deciding which button to press on each step, the wizard shows you the right one automatically:- Next when there's nothing to save- Save and Next once you've made changesThe result is one clear path forward through every step of setup — less clicking, less second-guessing, and fewer accidental "did I save that?" moments.This is available org-wide with no extra configuration required.📄 Learn more: Ticketing Configuration ---2. Chat Playback transcripts now land in ~10 minutesChat Playback lets you replay any user conversation with the AI Assistant to debug issues and troubleshoot with your partners. The biggest friction was the wait — transcripts used to take 3–4 hours to become available.Now, transcripts surface within ~10 minutes of a conversation completing.That means when a user reports an issue, you can pull up the session and see exactly what happened almost right away — including:- The full conversation replay: the user's question, the assistant's reasoning, its responses (text and cards), and citations- The detail behind each turn: plugin steps, plugin selection and execution, intent, and how the final response was generated 📄 Learn more: AI Assistant Chat Playback 

 Key Dates Frontier: June 29, 2026Standard: July 1, 2026Basic: July 6, 2026 To improve the user experience, we're updating how bold text works across responses on the web. Bold now serves as a structural tool only, appearing in two places: standalone headings when heading syntax isn't available, and short run-in labels at the start of a paragraph, followed by a colon.BeforeAfter  Learn more about our model upgrades here.

For customers in Controlled Availability, Agent Studio now supports folder-based access control and role-based permissions, so you can open up development to more teams without exposing sensitive connectors or putting live plugins at risk.What was the problem?Every developer in your Agent Studio tenant could see, edit, and publish every asset, including connectors to sensitive systems like Workday or ServiceNow. The only safe option was keeping Agent Studio locked to a small set of trusted engineers, which directly limited how fast teams could scale.What's includedFolders: Public (org-wide), Personal (drafts), and admin-created custom folders (e.g., "HR Integrations"). Assets inherit folder permissions automatically. Four roles: Manager, Developer, Operator, Viewer, assignable per user per folder or asset. Dependency enforcement: Running an asset requires Use permission on its full chain, including every action and connector it calls.  Boundary states: When blocked, the UI names exactly what's missing and surfaces a "Contact owner" prompt.How to get accessThis feature is in Controlled Availability for a curated set of design partners. Key limitations in this release: no group-based permissions (individual users only), flat folder structure only, no bulk admin actions, and MCP Servers are outside RBAC scope.Docs: linked here   

 🎉 Slack UX Improvements Now in Controlled Availability!We’re excited to announce that our new Slack UX Improvements package is now available in Controlled Availability.These updates introduce a set of interface enhancements designed to deliver a best-in-class Slack experience, improve trust in responses, and enable users to complete more work directly within Slack channels and DMs.In making agent interactions more transparent, your team can now interact with Moveworks in a far more intuitive, responsive, and visually rich environment.What is included in the Slack UX ImprovementsLaunch Native Global Actions, allowing Slack native popups directly within the chat interface to speed up task ​​​​Already included in the Slack UX RolloutOnce accepted into the Controlled Availability, your organization can validate Global Actions.These features will be deployed and validated with a small cohort of customers ahead of general availability. If your organization has already deployed Channel Resolver, you are an ideal candidate for this early phase.📝 Join the CA (if not already enrolled):Self-Nominate for Slack UX Improvements Controlled AvailabilityIf you have questions, ideas, or would like to showcase what you’re building, share them below - we’d love to hear from you!​ 

We are thrilled to announce that Enterprise Search Insights is now available to all customers who have Enterprise Search enabled for them.This dashboard fundamentally answers three questions : Who is using Enterprise search ?  How is Enterprise search being used, what are users doing on it ?  How to improve it ?   

We're excited to share that we are beginning to support GPT-5.4 in the Moveworks Reasoning Engine. We put GPT-5.4 through extensive internal testing before bringing it to you.Stronger Instruction Following. GPT-5.4 adheres more closely to the instructions that shape your plugin's behavior, resulting in more consistent, predictable outputs aligned to your configuration.Improved Cited Answers. When your assistant responds with information from your knowledge base, it now includes links back to the source documents far more consistently. Users get answers they can verify, not just bare claims.Improved Tool-Call Argument Filling. When your assistant calls a tool, it more accurately extracts the right information from the conversation. Whether it's a ticket title, a PTO date, or an employee ID, GPT-5.4 more reliably finds the particular values it needs.Why we’re so methodical with the upgradesBefore rolling out any new model, our team runs it against a suite of internal benchmarks. We then stage the rollout across tiers (Frontier first, followed by Standard and Basic) so that any issues surface in a controlled way before reaching your environment. It's how we make sure upgrades feel like improvements.When will GPT-5.4 be available in Moveworks?Target release dates by tier below:Frontier: Wednesday June 10, 2026Standard: Wednesday July 8, 2026Basic: Date TBDNote: Dates are for all accounts including GovCloud accounts. 🚧 Is there anything I need to do for this rollout? This change will be made to your plugin automatically during this period. There is no additional setup work required, nor any expected downtime.🛑 Are there any known issues? GPT-5.4's improved instruction adherence means that misconfigured or misaligned instructions in your plugin could have a more pronounced effect on behavior. If you see misaligned behavior, check the tool description, slot description, and the display instructions configurations within the plugin to more finely tune the desired response.