The Science Behind AI Classes
We are living in a very special time where a new technology generates as much fear as it does excitement: AI. Yet, we are increasingly using it, directly or indirectly, and achieving extraordinary results in our daily activities.
These results are evident in many fields, and education is certainly one of them. AI effortlessly simplifies many complex concepts to facilitate learning.
AI-Generated Videos as a Turning Point
Now, with the strong emergence of AI-generated videos, like Gen3 or Luma AI, we can take education to a higher level. Video is the most effective channel for transferring knowledge to us because it is the closest to reality, which is both auditory, visual, and dynamic, while also having the narrative power of storytelling.
Synthetic voices offer fluidity and clarity that are very difficult for most of us to achieve. They can also have different tones or accents that adapt to different populations. The generated videos even go beyond reality by enhancing it, giving it more strength, which facilitates memorization and ultimately the understanding of difficult-to-integrate knowledge.
Our brain records information based on its coherence, interest, and the impact it produces on a linear storyline. Therefore, if we combine the simplification capabilities of chatbots, the spectacular nature of generated videos, and clear synthetic voices, we can reach an optimum for learning any subject.
A Concrete Experience
This is the experience I conducted by creating two classes generated mostly with AI. I optimized them as much as possible to facilitate learning, taking into account a very broad audience.
To do this, I used a female avatar with a very clear synthetic voice that explains concepts created with ChatGPT in an educational, inspiring, and concise manner. The choice of a female avatar is due to the fact that the majority of tech enthusiasts are men, and learning is facilitated by a female representation.
The generated videos are both varied and impactful so that our brains can engage all areas and ensure good memorization. The structure is as inspiring as possible, alternating different emotions of fear and motivation, with a narrative logic to constantly maintain interest.
All these choices are based on human and social psychology to better help people understand with less effort.
How Were These Videos Created?
The first video addresses concrete solutions for any work involving AI. It is based on training material from a remote class I conducted for a Mauritian company in June 2024 and includes various tips, especially on chatbots. In the video version, I used several generated videos to illustrate some difficult passages, but within a traditional course approach.
The second video focuses on finance and uses many more generated videos to provide more inspiration to the audience. The narrative is deliberately generated with AI to have the strongest possible psychological impact. The average time between videos is 7 seconds to maintain a high pace.
More than 2,150 images were generated using Flux, from which 680 videos were created with Gen3 Alpha. Of these, only 80 were used in the final video. Additionally, 360 questions were asked to chatbots like ChatGPT or Claude. These numbers are important because despite AI’s interesting results, multiple samples must be generated to achieve convincing results for the general public.
This is just the beginning, and there will certainly be many videos in this format in the future because they have all the ingredients to surpass human-led education. The next step will likely mix learning with interactivity, where anyone can explore complex areas of knowledge with videos from adaptive conversational agents. The sky’s the limit!
Conclusion
AI-powered education is still in its early stages but has great potential, especially thanks to generated videos and synthetic voices. Two experiences were created to show two concrete examples: practical advice on benefiting from chatbots and solutions related to finance.
However, like any scientific social experiment, it could be a success or a failure, depending on public opinion.
Does this learning format seem relevant to you? Feel free to share your thoughts in the comments.
This story is published on Generative AI. Connect with us on LinkedIn and follow Zeniteq to stay in the loop with the latest AI stories.
Subscribe to our newsletter and YouTube channel to stay updated with the latest news and updates on generative AI. Let’s shape the future of AI together!
