Visual Content is an expansion of Yoodli AI Roleplays that enables experiential learning in a single session. During a roleplay, the AI can present visual materials live and guide the learner through them like a tutor, then transition into practice and provide immediate feedback.
What is Visual Content?
Visual Content is a type of roleplay where the AI actively presents visual materials such as slides, PDFs, and images during a live, conversational learning experience.
In this release, the AI can:
Present content like a human tutor (slides, PDFs, images)
Answer questions during the presentation
Adapt the flow based on learner interaction (skip a slide, go back, dive deeper)
Transition seamlessly into a full roleplay
Provide simple, contextual feedback at the end
Important clarification: Visual Content does not observe or analyze the learner’s screen. The AI is the one presenting. This is different from Yoodli’s existing “share screen” feature. Learners cannot share their screen with the AI while the AI is presenting.
With Visual Content:
A single Yoodli roleplay can teach the concept, let the learner practice it immediately, and then reflect with feedback, all in one continuous session.
How Visual Content supports Learn, Practice, Apply
LEARN
The AI presents admin-uploaded materials like a tutor. Learners can pause, ask questions, and go deeper into specific slides or concepts. The AI’s explanations are grounded in the uploaded materials.
PRACTICE
The session can transition directly into a full roleplay using the same content context. This removes “cold” roleplays that lack setup or context.
APPLY (DO)
At the end, the AI provides simple, contextual feedback:
What went well
What could be improved
Note: this does not replace Yoodli’s summary page and detailed feedback.
Feature specifications (Feb 12 launch)
Supported content
Slides
PDFs
Images
Videos (out of scope for Feb 12)
Core capabilities
AI understands uploaded content
AI presents content during a session
AI explains and clarifies on demand
AI can lead continuously or respond reactively
Learners can navigate via natural language
Session can progress from teaching → practice → feedback
Create a roleplay with Visual Content
Create a roleplay
Upload files under “Visual Content”
After upload, you will see individual slides that the AI will display, along with a brief description of each slide
Optionally rename slides and refer to those slide names in your scenario instructions for best results
FAQs
What makes this different from a normal roleplay?
The AI can teach the material first, then let the learner practice with the same context.
Can learners interrupt and ask questions?
Yes. The AI can pause, explain, and resume.
Is this replacing training content?
No. It turns existing content into an interactive learning experience.
Does the AI control the session?
Yes.
Do I need to upload the same file in Context for the AI to reference?
No. Uploading the same file is redundant. If a file is uploaded for Visual Content, the AI already has access to it. Add supporting materials that do not need to be presented, such as FAQs or a transcript of a prior call where the material was presented.
Is this the same as screen sharing?
No. An important distinction is that the AI presents content. The AI does not observe or analyze the user’s screen.
How does Visual Content affect rubrics?
Visual Content expands what you can evaluate. In addition to assessing conversation skills, your rubric can now measure how effectively the learner interacts with and uses the presented materials.