AI as a critical thinking coach
Can you use AI to promote critical thinking? That is what Edward De Vooght (Arteveldehogeschool) wonders. For the TITAN research project, he is developing AI tools that stimulate critical thinking and combat disinformation. Initial tests have already shown that artificial intelligence can promote reflection and critical thinking, but we are not used to an AI platform that asks questions instead of providing answers.
That you can use AI to work more efficiently and faster, and to look things up easily is obvious to many by now. In my work as a researcher at Arteveldehogeschool, I use ChatGPT to purposefully search the web, summarise it and reformulate it. But can AI also encourage thinking instead of just replacing it?
That is exactly the purpose of the European research project TITAN, which uses AI to increase critical thinking among European citizens to combat the threat of disinformation. For TITAN – a project funded under Horizon Europe – the Arteveldehogeschool, VRT and VUB work together with 11 other partners from Italy, Greece, Denmark, Finland, the UK and Bulgaria to develop AI tools that help citizens with critical thinking. Specifically, these include a coaching chatbot, which asks critical questions, and a discussion forum that gives you a comprehensive understanding of certain topics.
From March to May, we tested two of these AI tools with 22 higher education teachers and students, which resulted in a lot of insights but also revealed some challenges.
On the one hand, we learnt that AI has great potential to help people reflect and critically engage with disinformation. Students reported that the tools were very effective in helping them think more carefully about information they encountered online. They even found that the tools had a great impact on their critical thinking skills. While the coach helped them dwell on certain articles and think more deeply about the reasoning used, students were positive about the discussion forum, which, by organising arguments like a mind-map, provided a handy overview of their peers’ arguments and opinions.
On the other hand, we are spoilt these days by the luxury that platforms like ChatGPT offer us. The great ease with which ChatGPT can answer all our wild questions ensures that students are not very interested in a platform that asks questions (for now). It is counterintuitive to talk with a chatbot that mostly asks questions and offers few answers. The biggest challenge for a project like TITAN is being able to convince users that such tools are useful to them.
The TITAN research project runs until August 2025, giving us time to optimise these tools and to adapt them to users’ needs. Either way, it is already clear that AI is capable of supporting critical thinking in a personalised way, but that we ourselves have to want it first. So, the question is not whether AI can help us with critical thinking but whether we want and will use that help. The idiom goes: You can lead a horse to water, but you can't make it drink. But what’s more applicable in this case: You can fill a stream with water, but the horse might not want to go there.
Edward De Vooght
Edward De Vooght is een senior researcher aan Arteveldehogeschool gespecialiseerd in communicatie en kritisch denken.
Related courses
SAIAR Summer Studio: From latent to physical space
Zomerschool - Kortrijk - Howest Hogeschool, SAIARlab
HPC Café
Seminarie - Online, Karlsruhe - KIT, HammerHAI, Training geselecteerd in de Belgische AI Factory Antenna (AIFA)