El futuro del aprendizaje mediante vídeos: cómo hacer que los vídeos sean más accesibles, fáciles de buscar y atractivos

Video has become a permanent part of how colleges, universities, businesses, healthcare organizations, and government agencies teach and share knowledge. The COVID-19 pandemic accelerated the adoption of video, particularly in higher education and healthcare. The increase in video creation has led to an increase in expectations about video. A lecture recorded today may support the students in the room, a remote learner watching that evening, an employee searching for information six months later, or someone revisiting the material years from now. 

For much of the history of video learning, the technology challenge was getting that video captured, stored, managed, and delivered reliably. Those challenges still matter. Anyone responsible for a large video library knows that reliability, security, governance, and scale are not solved problems that can simply be ignored. 

The more interesting question now is what happens after the video has been captured. 

Research into educational video gives us some useful direction. Cynthia Brame’s review of the literature on educational video identifies three factors that contribute to effective learning: maintaining learner engagement, encouraging active learning, and managing cognitive load.[1] Those principles suggest that the future of video learning is less about accumulating more hours of content and more about helping people use those hours effectively. There is a fundamental transformation when a viewer moves from passively watching a video to actively engaging with the content. 

For Mediasite, I believe that means continuing to evolve from a platform that captures and manages video into one that helps learners find, understand, and engage with the knowledge contained within it. 

Help Learners Get to What They Need 

Some learning experiences benefit from short, focused videos. Others simply do not fit that model. 

A university lecture may run for an hour. A medical procedure, technical demonstration, employee town hall, research presentation, or professional training session may need even more time. Long-form video has an important place in learning. 

At the same time, we know that length affects how people engage with video. Philip Guo, Juho Kim, and Rob Rubin analyzed 6.9 million video-watching sessions across four edX courses. Their research found substantially higher engagement with shorter videos, with viewing engagement declining as videos became longer.[2] 

I do not think the lesson is that every piece of educational content needs to be condensed into a six-minute video. The lesson for video platforms is that we need to give people better ways to navigate longer content. 

If a student needs one concept from a 60-minute lecture, the best answer may be helping that student find the six minutes that matter. 

Chapters are one part of that. Searchable captions and transcripts are another. Slides, OCR, metadata, and strong descriptions all provide additional ways to understand what a video contains and whether a recording is relevant before pressing play.  

This is an area where AI can be useful without making AI the center of the learning experience. Speech recognition can make the spoken content searchable. Computer vision and OCR can identify information presented visually. Language processing can assist with topics, keywords, and organization. 

Generative AI creates additional possibilities, but it also deserves additional caution. A generated summary or answer is fundamentally different from a transcript of what an instructor actually said. Accuracy, provenance, privacy, institutional policy, and appropriate human review all matter. In education especially, convenience cannot come at the expense of trust. 

The goal should not be to insert AI into every interaction. The goal should be to help learners reach useful, reliable information with less friction.

Accessibility Needs to Be Part of the Workflow 

Accessibility is one of the clearest areas where video learning needs to improve. 

For many years, accessibility was approached as a remediation task. A video was created first. Captions and other accommodations were added later, sometimes only after someone requested them. Today, there are legal requirements and general expectations that video content is captioned. Other accommodations, such as Audio Descriptions or sign language are not as common. 

That model is increasingly difficult to defend, both from a learner experience perspective and from a regulatory one. Viewers have better outcomes when they can access and understand the content of your video, so creating accommodations is the right thing to do. We need to set a higher bar for ourselves than just reaching compliance and hoping to avoid legal exposure. 

There is also a strong research case for looking at captions as more than an accommodation. Morton Ann Gernsbacher reviewed more than 100 empirical studies and found that captions can improve comprehension, attention, and memory. The benefits extend beyond people who are deaf or hard of hearing to include people watching video in a non-native language and other learners.[3] 

The legal environment is also changing. 

In the European Union, the European Accessibility Act came into application in June 2025 for covered products and services. Among its provisions are accessibility requirements for services providing access to audio-visual media, including support for accessibility components such as subtitles for people who are deaf or hard of hearing, audio description, spoken subtitles, and sign-language interpretation when those components are provided with the media.[4] The precise applicability depends on the service and jurisdiction, but the direction is clear: accessibility is increasingly part of the expected digital experience. 

In the United Kingdom, public-sector organizations are already subject to the Public Sector Bodies Accessibility Regulations as well as broader obligations under the Equality Act 2010. Current UK government guidance directs covered public-sector websites and mobile applications to meet WCAG 2.2 Level AA. That affects many of the institutions that depend heavily on video for teaching and public communication, including most higher education institutions.[5] 

In the United States, public institutions are preparing for more specific requirements under the Department of Justice’s ADA Title II web and mobile accessibility rule. The rule generally requires covered state and local government web content and mobile applications to meet WCAG 2.1 Level AA. Following an April 2026 interim final rule, entities with populations of 50,000 or more generally have until April 26, 2027, while smaller entities and special district governments generally have until April 26, 2028.[6] The Department specifically notes that a state university uses the population of the state for determining its compliance deadline, rather than its student enrollment. Similarly, the U.S. Department of Health and Human Services has updated its Section 504 regulations to require recipients of HHS federal financial assistance to make covered web content and mobile applications conform to WCAG 2.1 Level AA. Following a 2026 extension, larger recipients generally have until May 11, 2027, to comply, while smaller recipients have until May 10, 2028.[7] 

Those dates may feel distant until an institution considers the size of its video library. 

A university with years of recorded lectures cannot approach accessibility the same way it would approach a handful of new web pages. Especially with the explosion in growth of video creation post-pandemic, institutions need a way to understand the accessibility of their libraries, identify the content that needs attention, prioritize work, and track progress. 

That thinking is behind capabilities such as the Mediasite Accessibility Dashboard. Instead of treating accessibility as a setting buried inside an individual video, institutions need visibility across their content. Which recordings have captions? Where are there accessibility gaps? What should be addressed first? 

Making that information visible gives institutions a better chance of managing accessibility as an ongoing program. 

Captions are also only one part of the experience. Audio descriptions, accessible transcripts, keyboard navigation, screen-reader support, visual presentation, and the accessibility of the player itself all contribute to whether someone can actually use the content. Additionally, offering user preference for things like theme (light, dark, high contrast), caption font/color/size/positioning, or language all make content more accessible. 

Accessibility works best when it becomes part of the normal video lifecycle.

Give Learners More Than One Way to Learn 

One of the advantages of digital learning is that the same source material does not have to produce the same experience for everyone. 

A learner may choose to watch a lecture from beginning to end. Another might follow along with the transcript. Someone may search for a specific concept and start halfway through. Another learner may need captions, audio description, or a translated version of the material. 

These are not necessarily separate pieces of content. Increasingly, they can be different ways of accessing the same knowledge. 

This aligns with the principles of Universal Design for Learning. CAST’s UDL framework encourages multiple means of representation, including alternatives for auditory and visual information, support for understanding across languages, and different ways to highlight important concepts and relationships.[8] 

Technology can make those alternatives easier to provide at scale. 

Automatic speech recognition has already changed the economics of captioning. Translation technology continues to improve. Audio description workflows are evolving. Search and content analysis can make recordings easier to navigate. 

The important measure is whether these technologies help people learn. An automatically generated output that is inaccurate, difficult to verify, or disconnected from the original material does not accomplish that goal. 

This is another reason to take a measured approach to generative AI. There are promising applications, particularly for assisting with time-consuming content enrichment. There are also legitimate concerns about hallucinations, bias, data use, intellectual property, and the loss of context when complex material is reduced to an AI-generated answer. 

Institutions should be able to make informed choices about where these capabilities are appropriate. Video platforms should provide transparency and control rather than assume every customer wants the same approach. 

Video Can Be More Active 

Finding the right information is only one part of learning. Learners also need opportunities to process it. 

Brame’s review of educational video research emphasizes active learning alongside cognitive load and engagement. Techniques such as embedding questions, connecting video to assignments, and giving learners more control over their experience can encourage deeper processing of the material.[9] 

This has implications for how we think about the video player. 

A player does not have to be a rectangle with a play button and a timeline. It can provide chapters, links, questions, supporting materials, discussion, and other ways for learners to interact with what they are watching. Analytics can then help instructors understand how content is being used, where learners disengage, and which material may need additional explanation. 

There is considerable room for innovation here that has nothing to do with generative AI. Mediasite offers the ability to add interactive quizzes, buttons, hotspots, and polls directly onto your videos to directly engage viewers. 

Sometimes the best improvement to a learning experience is simply better navigation, a well-placed question, an accessible transcript, or an easier way to return to an important concept. 

Unlock the Knowledge We Already Have 

Perhaps the biggest opportunity is sitting in video libraries that already exist. 

Many Mediasite customers have spent years building collections of lectures, research presentations, training programs, events, and institutional knowledge. Some have been doing so for decades. 

Those recordings represent an enormous investment. 

Their usefulness, however, depends on whether anyone can find what is inside them. 

A title, date, and presenter can tell us only so much about an hour of video. Once the spoken words, slides, visual information, chapters, and other context become discoverable, an archive becomes much more useful. 

Imagine a student being able to locate every place a particular theory was discussed across several courses. A researcher could find a presentation from years earlier without remembering its title. A new employee could find the exact section of a training session that explains a process. A faculty member could discover material already available in the institution’s library and incorporate it into a new course. 

There are technical questions to solve around permissions, governance, accuracy, and responsible use of AI. Those questions are important. They should shape how these experiences are built. 

They should not keep us from recognizing the opportunity. 

Where Mediasite Goes From Here 

Mediasite has spent more than two decades helping organizations capture, manage, protect, and deliver video. Those fundamentals remain essential. 

The next stage of video learning builds on that foundation. 

We can make content easier to navigate. We can provide better information about accessibility and give institutions tools to improve it. We can offer learners more ways to experience the same material. We can make video more interactive. We can use carefully selected forms of automation and AI where they provide real value while maintaining the trust that education requires. 

Most importantly, we can help institutions get more value from the knowledge they have already captured. 

The future of video learning will not be defined by how many hours of video an organization can store. It will be defined by how effectively people can find, access, understand, and use the knowledge inside those hours. 

That is the opportunity I see ahead for Mediasite.