At the AI Learning Centre, we have been trialling/will be trialling a current-content layer across a small number of courses, giving students timely access to AI-curated research and discipline-relevant developments.

The process starts with the teaching team—not the technology. Together, we agree the sources, scoring rubric, publication thresholds and course settings. Lecturers then review and refine sample outputs before anything is automated.

Once approved, the workflow publishes only items that meet those predefined controls. Teaching teams retain post-publication oversight: they can review live outputs and performance, refine the rubric, amend settings, remove individual items or pause the workflow. Teaching and technical staff are also briefed on its capabilities, limitations and escalation process for unsuitable content.

The aim is not simply to give students more content. It is to provide the right content, at the right moment, in the right course.

At a high level, n8n orchestrates the pipeline. Our Snowflake reporting layer identifies the relevant course and teaching period from Blackboard data feeds, while Blackboard applies its existing enrolment-based access controls. Candidates from approved sources, including licensed APIs and structured feeds, are deduplicated and scored by an LLM against the agreed rubric.

Before release, scores, summaries, source metadata and automated publication decisions are stored in Supabase, creating a traceable record for monitoring, post-publication review and rubric refinement. Using an approved Blackboard REST API integration, qualifying items are then published to the designated course location as announcements or course-content items—without requiring a separate student-facing application or Blackboard plugin. Built-in access, security and data-protection controls support the workflow.

Samples of the output format:

Cryptocurrency monitoring (currently implemented)

Pharma and news monitoring (to be implemented)

Tourism Industry (to be implemented)

These outputs turn newly published source material into concise briefings with summaries, key takeaways, relevance indicators and links to the original publications.

Initial informal feedback has been encouraging, particularly in relation to class discussion and connections between theory and current developments. We are now planning to formally evaluate the approach’s contribution to engagement and learning outcomes.

Each item includes clear attribution, a link to the original publication and a notice that it must not be redistributed. It is published only in the relevant password-protected Blackboard course, where access is limited to enrolled students.

This is AI supporting the lecturer and curriculum—not replacing them.

Where could a current-content layer add value in your course?