-
New Feature
-
Resolution: Fixed
-
Major
-
Future Dev
-
MOODLE_405_STABLE
-
MDL-80890-main -
-
6
-
Team Hedgehog 2024 Sprint 3.1, Team Hedgehog 2024 Sprint 3.2, Team Hedgehog 2024 Sprint 3.3
Provide a way for (primarily) students to use AI actions to interact with course content.
Students should be able to directly interact with AI actions while accessing the learning content in a course. This could be via a context menu that appears when a student highlights course content (see image below), or via other calls to action (CTA).
The first part of this will be UX design to outline the user journeys and UI elements.
While there will be a common CTA to access the available actions, each action will have their own workflows and interface. For example an action to translate the content to a different language may include a step for users to choose the language they wish to have the language translated into.
Actions available to the student will be dependent on the Actions provided by the provider plugins and admin configuration.
Inline with our AI Principles we will need to provide notification to users that they are using an AI enhanced service, and provide them a way to understand what this means.
Interactions should also be logged and reviewable so (for example) a student can review their conversations later for review, or a teacher can review what aspects of a course students are needing assistance with.
Figma Design (WIP):
Technical Spike Plugin: https://github.com/mattporritt/moodle-local_assist
Note:
Interacting with embedded PDF files is currently out of scope due to a technical limitation, with the way PDF files are embedded in LMS. Currently LMS uses native web browser support to display PDF files to the user. This prevents Javascript from interacting with the PDF which in turn prevents leveraging client side AI functionality. If PDF.js support is added to LMS then interacting with PDF files would be possible.