AI Technology & Data Use
What AI technology powers D2L Lumi Pro?
D2L Lumi Pro uses large language models (LLMs) provided through Amazon Web Services (AWS) Bedrock. It leverages Anthropic’s Claude models, including Claude Sonnet and Claude Haiku. D2L designs prompts and provides course content as input for the model to process. On certain capabilities, custom instructions extend AI output to incorporate user provided guidance that can alter tone, audience, style, and more.
Is customer data used to train the AI models?
No. Customer data is not used to train the AI models used by D2L Lumi.
How are the AI models trained?
The underlying models are trained by third‑party providers such as Anthropic. D2L does not participate in training these models and does not fine‑tune them using customer data.
Access & Configuration
Can D2L Lumi be enabled for specific courses or organizations?
Yes. Administrators can control where D2L Lumi Pro is available using configuration settings and tool permissions. The configuration variable d2l.Tools.GenerativeAI.isEnabled determines whether D2L Lumi is enabled within specific courses or organizations.
Content Import & Creation
What file types are supported by D2L Lumi Content?
Currently supported formats include Word (.docx) for importing pre‑written units and PowerPoint (.pptx) for generating learning units from presentations.
What are the two content import options?
Import Pre‑Written Unit: Used when content is ready to publish and should only be formatted and organized by D2L Lumi.
Generate Unit from Presentation: Converts presentation slides into a structured online learning unit following instructional design best practices.
Does D2L Lumi process hidden PowerPoint slides?
Yes. D2L Lumi processes all slides in a presentation, including hidden slides. It is recommended to remove slides that should not be included before uploading.
Content Remixing
Can instructors customize remix prompts?
Yes. Instructors can modify remix prompts to adjust how content is transformed. Some examples include simplifying language, highlighting key terms, or aligning content with specific pedagogical frameworks.
Is there a limit to how many times content can be remixed?
No. There is currently no limit on remix actions.
D2L Lumi Feedback
Where is D2L Lumi Feedback available?
D2L Lumi Feedback is available within the Assignments tool in Brightspace.
What data does D2L Lumi Feedback use?
D2L Lumi Feedback can draw information from learning outcomes, rubrics, course content, assignment submissions, and instructor annotations.
Can the autograde feature be turned on or off?
Yes. Autograde functionality is configurable and can be enabled or disabled.
D2L Lumi Tutor
Does D2L Lumi Tutor require D2L Lumi Chat?
No. D2L Lumi Tutor can be used independently of D2L Lumi Chat.
What knowledge does D2L Lumi Tutor use?
D2L Lumi Tutor references course content and linked resources rather than external internet sources.
Can institutions add additional knowledge sources?
Yes. Institutions can upload or sync additional resources such as academic policies, knowledge base materials, or advising resources.
D2L Lumi Quiz
What question types can D2L Lumi generate?
D2L Lumi can generate multiple choice, true/false, short answer, written response, and multi‑select questions.
Can D2L Lumi generate feedback for quiz questions?
Yes. D2L Lumi can optionally generate feedback for multiple choice, true/false, and multi‑select questions.
Free Trial
Is there a free trial available?
Yes. D2L offers Lumi TryNow experiences that allow organizations to explore D2L Lumi capabilities before purchasing.