Home
News
Learning Technologies 2025

Learning Technologies 2025

Over the past few days, we attended the Learning Technologies 2025 event in London, which is dedicated to companies that use technology to support training and knowledge sharing in professional settings.

This is an event we’ve been eagerly awaiting because it addresses the technological areas we’ve been focusing on in recent months—namely, understanding the innovative features that can be integrated into digital training platforms, with a particular focus on artificial intelligence.

And it was precisely in the field of AI that we had the highest expectations: when it comes to artificial intelligence, the line between innovation and the mere rehash of well-established technologies can be very fine.

We'll tell you how it went.

Platforms

From the very first posters and booths, artificial intelligence was in the air: from product names to technologies, from platform features to courses, the acronym AI was present in at least 90% of the event. But as we talked to people, asked questions, and gathered information, we realized fairly quickly that even in this context, a mistake—one that has become quite common in recent months—is being made: using the term AI as a synonym for ChatGPT (or similar tools).

In fact, the initial impression was that most industry professionals, feeling compelled to use artificial intelligence, had integrated ChatGPT, Gemini, or other tools—more to respond to an apparent growing demand than to actually improve the platforms’ functionality.

What we've seen in most cases has been a port of OpenAI's generative AI to the platform, without providing any real added value.

But this has brought up once again one of the big questions we’ve been asking ourselves over and over again in recent months while working with AI on digital training platforms: How do you manage internal security?

Cybersecurity and data protection are topics that are somewhat less in vogue than artificial intelligence, but they remain the cornerstones of technological innovation, especially in the business world. Communicating with models like ChatGPT means putting sensitive information online; the fact that a model is integrated into a platform does not automatically protect the shared data.

When we asked how this issue was being addressed—given that many of the platforms we looked at emphasized data security, at least in their slogans—we were unable to get truly satisfactory answers . “Data is protected.” Yes, but how?

Contents

Since we couldn’t find a clear answer to that question, we shifted our focus from platforms to content. The world of digital education consists of technological tools and course development; it is therefore particularly interesting to examine companies that specialize in content production to see what kinds of innovations they are offering.

Content production can certainly be significantly accelerated compared to traditional models: from graphics to text, from translations to videos, there are many areas where artificial intelligence can provide a major boost. This translates into immediate cost savings: comparisons and discussions have shown that savings in terms of resources used for production can reach as high as 70–80%.

From the perspective of technological innovation, attempts to use augmented reality to create interactive courses are also interesting, particularly in certain sectors: a prime example is machinery maintenance, which can be practiced during training in a virtual, assisted environment and then applied in real-world situations. However, we would say that, in general, we are still in the early stages of using this type of technological integration in the business world.

Numbers

Corporate training is, among other things, a means of improving business performance. Being able to quantify these improvements is, at least in our view, one of the key goals to be achieved through the intelligent use of technology.

In this regard, we found the analyses by Thrive and Coursera particularly interesting; they estimated that for every 1% invested in corporate training, there is an incremental return of 4.7% on revenue. This increase stems not only from the use of AI tools, but also—and above all—from understanding them: understanding how agents are used, what a machine learning process is, what training entails, and so on.

Considerations

Learning Technologies 2025 was an interesting event that provided us with some insights and several important confirmations.

Functionality. Security. Integration.

First and foremost, we remain convinced that artificial intelligence must serve a specific purpose and should not be merely a trend. Any technological integration must serve a purpose, whether it be an improvement or an innovation. As mentioned, from our perspective, an extremely interesting aspect right now is the correlation between business data and training data; in this regard, artificial intelligence can serve as a catalyst and an extremely valuable analytical tool.

All of this must be developed with the utmost attention to corporate security, keeping most of the data offline. This is a fundamental consideration for us, but given the current trends in the market, it seems to us that not all operators are giving this aspect the attention it deserves.

Finally, there is the issue of integration—another term that is often used as a buzzword but lacks real technological support: most of the platforms we’ve seen integrate with the major CRMs and ERPs, while integration with customized systems becomes more complex.

These are our thoughts on LT 2025. In the coming weeks, we’ll keep you updated on the latest news directly related to the Viasky world, which you can find on our website and blog.

Image by Fulvio Bottoni

by 

Fulvio Bottoni
Chief Technical Officer
Share this post:

More posts