In 2018, EDUCAUSE Review ran a piece called Universal Design for Learning and Digital Accessibility: Compatible Partners or a Conflicted Marriage? It surfaced something most of us already know but just don’t say it out loud: captioning workflows tend to follow a “good, fast, cheap, pick two” logic, and the picking usually happens under compliance pressure rather than pedagogical reasoning.
The accessibility team owns the budget line, if there is one. The standard is “accurate enough to meet WCAG.” Once that bar is cleared, attention moves on.
Fast forward to today, where captioning quality is a lot higher, skewing that previous pick-two logic. AI generated captions have closed much of the accuracy gap, and transcripts are now genuinely searchable at scale. That changes what captions are for.
Practitioners are reporting students using transcripts to locate specific concepts, skim rather than rewatch, and return to particular moments in a recording. The research base on this is still thinner than the practitioner consensus, so I’d hold the claim loosely, but the pattern is worth taking seriously.
Which raises a question I’d like the community’s view on. If captions function pedagogically for many students, are we resourcing them in the right place, to the right standard, by the right people?
A open questions:
1. Where does captioning sit in your institution: accessibility, learning technology, or the academic’s own desk?
2. Do your students use transcripts as a study tool? How would you actually know? Has anyone tried to find out?
3. Are you seeing any trends within your organisation to increase captioning if you are already captioning content?