From Training to the Real Thing: What's New at TreadEd

With our new release, learners can show what they know and speak freely instead of typing. Creators get an agent that keeps the whole learning path in view.

A few months ago, I described TreadEd here as an “ergometer for the mind”. An AI assistant takes resistance away from us; a good tutor gives it to us, exactly as much as we need in order to learn.

But resistance alone doesn't motivate anyone. Nobody gets on an ergometer because they enjoy sweating. We train toward something, and we want the effort to land where it makes a difference.

That's what our new release is about. Two new features make learning more motivating: diagnostics give learning a goal, and voice mode puts the effort into thinking instead of typing. The third is for the people who build learning paths: the workspace agent.

An exam situation causes stress when target > actual, or when it's unclear how target and actual relate to each other (target ? actual).

Practice exams can give learners a clear sense of whether actual = target. And if not, they can turn an actual < target into an actual → target.

That's exactly what diagnostic steps are for. With them, learners get a target to work toward, and the learning steps get the valuable actual they need to close gaps in a targeted way.

It's as if your teacher graded those practice exams so you could then work on exactly the gaps that remain.

The new voice mode lets learners have a spoken conversation with the tutor. Switching back to the familiar writing mode is always possible.

On small screens in particular, this frees up the view and, among other things, makes labs usable on mobile devices.

And even without a lab, voice mode can work like an interactive podcast: a conversation with an expert in which learners have to think along actively.

Every answer is information about the actual, and with narrowing and optional criteria, intelligent adaptive follow-up questions can map a whole cluster of relevant criteria precisely and enrich the learning process that follows.

One principle matters to us here: Results belong to the learners. They stay in the learner's profile and are shared only if learners actively choose to: with the provider of the content, either the result alone or together with the conversation, and revocable at any time. If someone disagrees with an assessment, they can file an objection, and the host takes a look.

Creating should feel free, and still everyone should get the orientation they need. The workspace agent turns the potential of AI into an intelligent companion that supports you freely as you create: with an eye on the whole, and with the tools and the freedom to turn exactly the dials you're working on right now.

Just as voice mode gives learners more freedom to think, the workspace agent gives you, our creators, more freedom to design.

Existing formats remain compatible, of course, but diagnostic steps can only be built with the workspace agent.

If you have questions, get in touch with us directly. We firmly believe these new possibilities will make your goals even easier to reach, and we look forward to hearing from you.

About the author

Dr. Christoph Kenntemich is Co-Founder of TreadEd.

TreadEd is a creator tool that makes it possible to design dialogue-based learning paths in which learners are always optimally supported through one-on-one guidance by an AI tutor. Experienced employees fully shape the tutor didactically, passing on their knowledge and experience at scale.