E-learning: generative AI at the service of learner engagement

The two main challenges of e-learning for learners are the transmission and The commitment.

The two are very often linked because an engaging course is also an interesting course in which the learner discovers new concepts useful for his work.

Unfortunately, most e-learning courses have a sad reputation for being boring and not very useful.

In summary, creating engaging and effective content mainly requires two things:

• Skills in educational engineering
• time, lots of time.

“Hire me if you can”

Humans need to contextualize information in order to remember it. This is due to the structure of our brain that works by connecting neurons together to remember concepts.

So one of the keys to e-learning is Learner engagement in relation to training.

The educational content must therefore adapt to this constraint and It can't just be texts containing the information you need to remember.

Chez Didask, our researchers and educational engineers have understood this, which is why our platform offers various types of rich content for learners:

• Multiple choice question
• Debate
• Situation
• Categorizer
• Interactive image
• and others!
  ‍

Each of the content types has been designed specifically for Captivate learners' attention AND maximize learning.

By following this “learning framework”, course creators can design engaging courses that can actually transmit knowledge to learners.

But of course, creating an imaginary debate between two protagonists in order to deconstruct a received idea takes much more time than simply posing the information as it is.

With this challenge in mind, we decided to turn to AI to facilitate the creation of educational content offered on Didask.
‍

Beyond Prompt Engineering, the LLM Engineering

Generative AIs are great tools for generating content but their main flaw is that by definition, they only generate text.

Luckily, we can use Prompt Engineering techniques to ask the AI to generate something other than text. We're going to ask it to generate structured text.

Structured text can be used in computer programs such as The Didask platform. This is what developers deal with on a daily basis in their work.

Once we are able to ask the AI to generate structured data for us, we will be able to use them in the Didask platform to create business entities representing educational content.

This discipline combining Prompt Engineering and Software Engineering is called The LLM Engineering And she is the one who massively increases productivity people thanks to generative AI.

Knowledge transfer

Although the abilities of generative AIs are very far from those of a human brain, I propose a bit of anthropomorphism in order to better understand how they can help us in our daily lives.

When our Prompt Engineers are working on Didask educational AI, they write instructions to transform raw content into content with high educational value.

These instructions are very similar to instructions that could be given to a person with no knowledge of educational engineering in order to create a training course.

Example: a prompt whose purpose is to generate a MCQM-type exercise :

From a concept that learners must remember, generate a question of intermediate difficulty as well as 3 possible answers. Wrong answers should be based on preconceived ideas about the concept to be learned. The answers should not contain any clues to determine their veracity.

At Didask, our educational engineers and researchers in cognitive sciences have written dozens of instructions of this kind using a tool that allows the conditional sequence of instructions in graphs. (Rivet)

These sequences of instructions can be considered as a manual for creating content with high educational value, which can be used automatically by a generative AI (I also talked about it in my previous article) Beyond ChatGPT: Generative AI at the service of pedagogy).
‍

Thanks to this “manual”, anyone is able to create content with high educational value by being assisted by our AI.

The best of co-pilots

Educational engineers are also using AI to assist them in the generation of educational content. Thanks to their know-how, they are able to give additional instructions to the AI in order to guide the generation.

The great added value of AI is that it allows them to test different types of content or situations more quickly to teach a concept.

And therefore to create ever more qualitative training courses in record time.

For example, our team of educational engineers multiplied by 3 the number of modules created without recruitment and without additional hours of work, thanks to the use of Didask educational AI.

The modules generated in this way remain of very good quality thanks to the transfer of skills that has been carried out to educational AI by our engineers and researchers.

So don't wait any longer and come and try the future of the learning experience 🚀

‍

‍

To continue reading, I recommend the following articles: On commitment:

• Digital learning: 5 golden rules to disengage your learners by Benjamin Poucin
• Why EdTechs need to distinguish engagement and learning by Son Ly

On educational artificial intelligence:

• Adaptive learning and Generative AI: the winning combination for online training! by Svetlana Meyer
• AI & eLearning design: I compared ChatGPT and Didask... the verdict is final! by Benjamin Poucin
• How can AI further accelerate the transfer of skills? by Svetlana Meyer

On the transmission of skills:

• What does pedagogical effectiveness mean? by Svetlana Meyer
• E-learning: putting an end to the myth of the miracle modality to accelerate the development of skills by Benjamin Poucin
• 3 essential keys to move from memorization to application in eLearning training by Svetlana Meyer