UCL AI Accessibility Agent

An AI-assisted workflow supporting alt text generation for Moodle learning materials - built within the standard Microsoft 365 Copilot experience and deployed organisation-wide, developed and launched jointly with the UCL Digital Accessibility team.
Microsoft 365 Copilot agent interface showing the AI accessibility agent configuration for UCL STEaPP - designed to support alt text generation for Moodle course images

Microsoft 365 Copilot · Microsoft 365 · Moodle · Accessibility · WCAG 2.2 AA · UCL STEaPP · Staff Guidance · Human Review

Live AI accessibility workflow deployed at UCL STEaPP
M365 Copilot Built within the standard M365 Copilot experience on a standard licence
Org-wide Deployed across UCL, not just STEaPP
WCAG 2.2 AA Accessibility standard the workflow supports
Microsoft Copilot Microsoft 365 Moodle

Project

I developed this AI-assisted accessibility workflow in collaboration with the UCL Digital Accessibility team - with their full involvement in design, scope, responsible use guidelines and approval. The project connects accessibility compliance, practical AI adoption and the everyday realities of how academic content is produced at scale.

The problem: digital learning materials across a portfolio of online postgraduate modules contain images, diagrams and screenshots that require accurate alt text under the Public Sector Bodies (Websites and Mobile Applications) Accessibility Regulations 2018. Alt text was inconsistently produced - sometimes technically absent, sometimes present but unhelpful. Manual retrospective auditing does not prevent the problem recurring. The goal was to support accessible content production at the point of authoring, not after the fact.

The workflow was launched jointly at the UCL Teaching and Learning Forum, where it was presented to colleagues across UCL as a model for responsible AI adoption in accessibility practice.

What I Built and My Role

  • Helped translate the accessibility need into a practical, governed workflow for academic staff
  • Designed and built the agent using the standard Microsoft 365 Copilot agent experience - navigating publication channels, tenant approval, user permissions, authentication, knowledge sources and licensing constraints to deliver an organisation-approved tool on a standard M365 licence
  • Worked with the UCL Digital Accessibility team on scope, responsible use guidelines and appropriate guardrails
  • Developed staff guidance on responsible use, including the expectation that AI-generated alt text requires contextual human review before use in live learning materials
  • Supported testing, launch and communication activity, including presenting the workflow at the UCL Teaching and Learning Forum

Note on tooling: this agent was built using the standard M365 Copilot agent experience - not Copilot Studio (a separately licensed premium platform). Working within standard licensing required carefully navigating agent distribution, tenant approval, permissions, authentication and knowledge sources. I have since gained access to Copilot Studio and am building on this foundation with its fuller capabilities.

Core workflow

  • Staff upload an image directly to the agent
  • The agent analyses the image and produces a suggested alt text description, formatted and ready to paste into Moodle
  • Explicit guidance is built into the workflow: AI-generated descriptions must be reviewed in educational context before use
  • Staff guidance explains what makes alt text useful - not just technically present but educationally meaningful
  • Human review is built into the workflow by design, not offered as optional
Microsoft 365 Copilot agent interface showing the AI accessibility agent built for UCL STEaPP - designed to generate alt text descriptions for images in Moodle course materials

Tools

Microsoft 365 Microsoft 365 Copilot Moodle LMS WCAG 2.2 AA accessibility guidelines UCL Digital Accessibility guidance Human review process Staff guidance documentation Responsible AI frameworks

Outcome

The AI accessibility agent is live at UCL STEaPP and shared organisation-wide. Staff guidance materials and a responsible AI use framework have been published. The workflow makes it easier for content authors to produce a meaningful first-draft alt text description, reducing the barrier to accessible content production without removing human oversight.

Moodle course page with browser accessibility inspector open showing colour contrast ratio 15.48 passing WCAG AA - illustrating the accessibility testing approach applied to Moodle content

WCAG AA colour contrast testing during Moodle development - accessibility built into content authoring, not audited afterwards.

Stakeholders

  • UCL Digital Accessibility Team - central partners in design, scope, review, approval and ongoing guidance development
  • Academic content authors - primary users producing Moodle learning materials
  • Programme teams - responsible for overall quality and accessibility of online modules
  • Learners - particularly disabled learners who rely on accurate alt text to access learning materials
Related work
Skills demonstrated

Responsible AI adoption M365 Copilot agent development WCAG 2.2 AA Staff guidance development Accessibility compliance Stakeholder collaboration Human-centred workflow design