AI & learning – we tested ChatGPT!

Accueil » Teach on Mars News » AI & learning – we tested ChatGPT!


In the spotlight for a few years now, artificial intelligence (AI) is asserting its position in the future of digital training: ChatGPT, YouChat, NotionAI and many other discussion interfaces have gained notoriety lately. Already affected by these changes, the digital training industry is directly impacted by easy access to powerful AI. Occasionally criticised, some see it as an opportunity, while others highlight the risks associated with over reliance on AI.

Temptation proved too great for us Martians – as patrons of futuristic technologies. So we gave it a go.

Discover the result of our Test & Learn on ChatGPT, as part of a discussion about our core business – digital training!

First question: Is there room for artificial intelligence in the digital training industry?
ChatGPT :

“Yes, artificial intelligence (AI) can be a major opportunity for the digital training industry. There are several reasons for this:

1. Customised learning: AI can help to customise learning for each learner based on their needs and level of expertise. By using data analysis, AI can create personalised training pathways which are adapted to individual learners.

2. Adaptive learning: AI can also be used to create adaptive learning systems which adjust in real time according to learners’ answers. Learners can thus benefit from more efficient and faster learning.

3. Learning assessment: AI can help to assess learning using tools such as voice and face recognition. This can help to give specific comments and to improve learning.

4. Real-time assistance: AI can be used to offer real-time assistance for learners. For example, chatbots which use AI can provide immediate answers to learners’ questions, thus enhancing their overall learning experience.”

Second question: Are any risks involved with over reliance on AI in digital training?
ChatGPT :

“Yes, there are risks involved with over reliance on AI in digital training, in particular:

1. Algorithmic bias: If data used to train algorithms is biased, AI can reproduce these biases in training results. For example, if the algorithm is trained using data which under-represents certain populations, it can reproduce this bias when customising learning.

2. Loss of human contact: If AI is over-used to provide training, learners may have less interaction with trainers and other learners. This can lead to a loss of human contact and collaboration, which can be essential for efficient learning.

3. Lack of personalisation: Although AI can be useful for personalising learning, it can also lead to excessive standardisation of training. If AI is over-used it can ignore certain aspects of learning which cannot be easily measured or quantified, such as learners’ personal interests.

4. Technological dependency: AI requires investments in technology and infrastructure, which can lead to technological dependency and high costs. Furthermore, if AI systems are not regularly updated or maintained, they can become obsolete and lose their effectiveness.”


Third question: Will AI substitute today’s trainers?
ChatGPT :

“AI will probably not totally substitute current trainers, but it could provide them with additional support and change their way of working. AI is capable of handling and analysing large quantities of data, which can be useful for trainers. AI algorithms can identify learners’ needs and adapt training content according to their performance and answers. Trainers can thus use this data to better grasp learners’ needs and personalise their instruction.
However, AI cannot totally replace trainers’ social and pedagogical skills. Trainers are able to forge relationships of trust with learners, understand their emotional needs and offer encouragement and constructive remarks. Trainers can also facilitate group discussions and encourage collaboration. These skills are essential for successful learning and cannot be substituted by AI.”

>>> Decryption

From this discussion with ChatGPT, we learn that AI will be able to:

  • help to improve learners’ learning experience,
  • provide them with more personalised training,
  • help with evaluation to assist them in real time,
  • offer support to trainers by providing them with additional data and analyses.

That said, the human dimension and support for trainers remain key for high-quality teaching and a successful learner experience.

At Teach on Mars, we support these innovations at the service of learning. And like any innovation, AI will be moulded by the desires and aspirations of those who use it. We have to pave the way for tomorrow’s tools, while giving everyone the time they need to become familiar with them.