Frequently asked questions
Short, lookup-friendly answers to the questions members and admins ask most. Use the headings to jump to a topic. Where a fuller walkthrough exists, the answer links to it.
Getting started
Section titled “Getting started”What is SUPERWISE Chat? A secure, multi-user AI chat workspace for your organization. You can ask questions, draft and analyze documents, search the web, work from your own uploaded knowledge, and collaborate with teammates — all with built-in governance so your organization can trust the answers.
How do I sign in? Through your organization’s single sign-on. You don’t create a separate password. See Sign in.
Do I need to install anything? No. SUPERWISE Chat runs in your web browser. There’s nothing to download.
Where do I start once I’m in? Type a question in the box at the bottom and press Enter. See Send your first message and the Two-minute tour.
Is there a mobile app? SUPERWISE Chat works in a mobile browser. A dedicated app is not part of the current product.
Sending messages
Section titled “Sending messages”How do I send a message? Press Enter to send and Shift + Enter for a new line. You can switch this to “Shift + Enter to send” in Settings → Preferences.
Is there a length limit on what I can type? Yes — a single message can be up to 4,096 characters. A counter appears when you’re getting close to the limit.
What are the buttons under a response?
| Button | What it does |
|---|---|
| Explain more | Asks for a more detailed explanation |
| Refine this | Asks the assistant to improve its previous answer |
| Try different approach | Re-answers your question a different way |
| Compare | Shows two versions side by side so you can pick |
| Copy | Copies the response |
| Retry | Regenerates the response |
| Search the web | Appears when web search wasn’t used; re-runs your question with the web on |
Can I edit a message after I send it? Yes — hover your own message and choose Edit. The assistant re-answers from your edit.
Can I stop a response while it’s writing? Yes. The Send button becomes a Stop square while a response is streaming.
Modes and speed
Section titled “Modes and speed”What are Interactive, Thinking, and Research? They’re response modes you pick in the box before you send. They trade speed for depth.
| Mode | Best for |
|---|---|
| Interactive | Fast, direct answers |
| Thinking (default) | Everyday questions that need reasoning |
| Research | Deeper questions that benefit from search and citations |
| Thought Partner | The most thorough mode — deep research and multi-step work |
See Modes explained.
Why does the answer sometimes show steps like “Understanding your intent”? That’s the thinking view — a plain-language look at how the assistant worked through your request. You can expand it to see timings, or ignore it. See Different from ChatGPT.
Working with your documents
Section titled “Working with your documents”Can I upload a file? Yes. Use the + menu or the paperclip, or drag and drop a file into the message box. See Upload a file.
What file types are supported? PDF, Word (DOC/DOCX), Excel (XLS/XLSX), PowerPoint (PPT/PPTX), CSV and TXT; data files like JSON, XML, YAML, Markdown and HTML; common code files; and images such as PNG and JPEG.
How big can a file be? Up to 100 MB per file.
What’s the difference between attaching a file and adding it to Knowledge?
| Attachment | Knowledge collection | |
|---|---|---|
| Lives | In one message | Across conversations |
| Reused | No | Yes — the assistant can cite it any time |
| Shared | No | Optionally, with teammates |
You can promote a ready attachment into a Knowledge collection in place.
What is a Knowledge collection? A set of documents the assistant can search and cite. Collections can be Personal (only you), Folder (a project team), or Tenant (everyone in your organization). Personal collections stay private — even your admins can’t read them.
Can the assistant read a public website?
Yes — type /crawl followed by one or more URLs, or use Add Source → Web Crawl on a collection. Private or internal addresses are blocked for safety.
How do I know where an answer came from? Grounded answers carry numbered citation markers and a Sources Used card. Click a marker to see the source title, snippet, and link. If your collection had nothing relevant, the assistant tells you it’s answering from general knowledge.
Web search
Section titled “Web search”How do I turn web search on?
Toggle it in the + menu, use the Web pill in the message box, or type /web-search on. State is remembered per conversation. See Search the web.
Why did an answer suggest searching the web? When the assistant answered without the web, it offers a Search the web button so you can re-run the same question with live results.
Organizing your work
Section titled “Organizing your work”How do I keep conversations tidy? Rename (double-click the title), pin the important ones to the top, and archive the rest. Archived chats are restorable from Settings → Archive. See Organize conversations.
What is a Folder / Project? A workspace that groups related conversations, an assistant setup, and shared knowledge. New chats you start while a folder is open join that folder automatically.
Can I search inside a conversation? Yes — the magnifying glass in the header, or Ctrl/⌘ + F, finds text and jumps between matches.
Can I save a good prompt to reuse?
Yes — saved prompts (templates) appear in the / menu when your organization has them enabled. See Saved prompts.
Collaboration
Section titled “Collaboration”Can I work with teammates in the same chat? Yes. Share a conversation and you’ll see each other’s messages live, with names and avatars. There are also Channels (persistent group spaces with threads and pinned messages) and Direct Messages with teammates — the assistant is available in all of them.
Can I share a conversation outside the app? You can export it (below) or share it with named teammates inside the app.
Saving and exporting
Section titled “Saving and exporting”Can I export a conversation? Yes — from the conversation menu, export as PDF, Markdown, plain text, or JSON. See Export your work.
What are Notes and Snippets? Notes are a place to keep useful text. From a response you can Save Snippet (when your organization has snippets enabled) to keep an answer for later, and turn it into an editable note.
Memory and personalization
Section titled “Memory and personalization”Does it remember things about me? It can. Personal memory captures preferences and facts from your conversations to make future answers more relevant. You review and edit proposed memories before they’re saved, and personal memory is private to you. Clear it any time in Settings → Account.
Can I change the look and behavior? Yes — Settings → Preferences covers theme (light/dark/system), sidebar density, send behavior, notifications, and more.
Can I talk to it instead of typing? Voice input and spoken responses exist but are off by default. If your organization turns them on, you’ll see a microphone in the message box and voice options in Preferences.
Trust, privacy, and safety
Section titled “Trust, privacy, and safety”Who can see my conversations? By default, your conversations are yours. Shared conversations, channels, and DMs are visible to the people you include. Your organization is separated from every other organization on the platform.
Can my admin read my private chats and Personal knowledge? Personal Knowledge collections are strictly private — admins do not get access. For what admins can and can’t see in general, see Your privacy.
Is my data used to train AI models? See your organization’s Data processing terms and the Security statement for how data is handled.
Why did it decline to answer or change my message? SUPERWISE Chat checks requests and responses against your organization’s safety and compliance rules. When something is held back, you’ll see a short notice explaining it. See Why it said no.
What shouldn’t I paste into chat? Follow your organization’s policies; as a rule, avoid sharing secrets you wouldn’t put in an internal document. See What not to share.
Account and limits
Section titled “Account and limits”Why am I being told I’ve hit a limit? Each organization has usage allowances. If you reach yours, new AI requests pause until the limit resets or an admin raises it. Check Settings → Usage.
How do I report a problem or get help? Use your organization’s internal support channel. Admins can find diagnostics in the Admin troubleshooting guide.
For administrators
Section titled “For administrators”How do I add people and assign roles? Through your identity provider and the admin tools. See Users and roles.
What roles are available? Roles range from Viewer (read-only) and Member (full everyday use) up through project and tenant administrators, plus focused roles like Channel moderator and Governance manager (read-only oversight: view traces, view policy alerts, export audit data). See Users and roles.
How is each organization kept separate? Every request is scoped to its own tenant, and one organization can never reach another’s data. See Tenant isolation.
Which identity providers are supported? Any standards-based (OIDC) single sign-on provider works, including the platform default and Auth0. Your existing SSO and user directory carry over. See Identity and SSO.
Can I see an audit trail? Yes. Role changes, settings changes, and governance decisions are recorded; settings and role-change logs are tamper-evident. See Audit and usage.
How do I set policies and guardrails? Through governance settings and per-assistant guardrail configuration. See Governance and policy.
How long is data kept? Retention is configurable per organization, with defaults for conversations, traces, and logs. See Data flow and logging.
How do I handle a data-subject request or deletion? See Data subject requests and deletion.
Related references
Section titled “Related references”- Modes explained — the response modes in detail
- Keyboard shortcuts — work faster
- Glossary — terms used across the docs
- The surfaces — how chats, folders, channels, and notes fit together