Generating content

Ask the Knowledge Assistant to produce documents, images, audio, and video, or search the web — JoySuite picks the right model for you.

Draft — pending review. This article was drafted from JoySuite's product marketing and has not yet been verified against the shipping product. Feature names, screens, and steps may differ.

The Knowledge Assistant does more than answer questions — it can also produce finished content. Ask for a document, an image, audio, or video, or ask it to search the web, and it returns a result you can refine and share. You describe what you want in plain language; JoySuite handles the rest.

You never pick a model

Different AI models are good at different things — one is stronger at reasoning, another at writing, others at images, audio, or video. JoySuite draws on several leading models behind a single interface and selects the best one for each request automatically. You never see model names, never switch providers, and never have to decide which AI to use. Just ask.

Because model selection is automatic, the same request works for everyone on your team — no configuration and no AI expertise required. As new models are evaluated and rolled out, your requests benefit without any change on your side.

What you can generate

Describe what you need and the Knowledge Assistant returns it in the right format.

  • Documents — drafts, summaries, reports, and presentations, including Word, PowerPoint, Excel, and PDF formats.
  • Images — a picture that matches your description, without a separate design tool.
  • Audio — voiceovers and other spoken audio from a script.
  • Video — generated video from a description or script.
  • Web search — answers grounded in current information from the web, rather than only what a model was trained on. This can include image and video search.

Generated content is a starting point, not a final deliverable. Review and refine what the assistant produces before you send or publish it — especially anything customer-facing or policy-related.

Grounding results in your content

Generation works best when it draws on your own material. Point the assistant at your knowledge sources — files, FlexDocs, and connected systems — so a generated report, summary, or presentation reflects your data and voice rather than generic content. For grounded answers that cite their sources, see Assistant responses.

If you generate training material, you can carry it into Course Studio to build and publish a course. And when a generation request is one you make often, turn it into a reusable custom command so your whole team can run it consistently.

Coming soon. A deep-research capability that pulls from multiple sources into a structured synthesis is planned but not yet available. Until then, use web search for current, web-grounded answers.