AI in education: Supporting learning without replacing thinking

For generations, education has been based on the idea that students develop knowledge and skills through active learning. Reading, practising, making mistakes and working through problems are considered important parts of the learning process. The growing use of artificial intelligence (AI) tools is changing how students approach these activities and raising questions about the role of independent thinking in education.

Students today can use tools such as ChatGPT, Gemini and Claude to summarise chapters, explain difficult concepts, generate written content, solve equations and practise for examinations. When used appropriately, these tools can make learning resources more accessible. A student struggling with a concept can seek another explanation, work through examples or ask follow-up questions at a convenient time.

However, there is a difference between receiving an answer and understanding how it was reached. Learning often involves effort: students consider a question, recall information, make mistakes, review their approach and try again. These activities help them develop understanding and apply knowledge independently. When AI provides an immediate response, some of these learning steps may be bypassed.

Efficiency, therefore, does not always translate into deeper learning.

A calculator offers a useful comparison. It can perform calculations quickly, but students are still taught basic arithmetic to develop an understanding of numbers and mathematical operations. Similarly, if a student asks an AI tool to write an essay before developing an argument independently, the final piece of writing may be well structured without necessarily reflecting the student’s own understanding.

This presents an important question for schools and universities: rather than focusing only on restricting AI use, how can educators ensure that students remain actively involved in learning while using these tools?

A complete ban on AI may not prepare students for workplaces where such technologies are increasingly being used. Instead, education can focus on developing AI literacy. This includes understanding when AI tools may be useful, recognising their limitations, checking the accuracy of their responses and evaluating the information they provide. Students should also learn when a task requires their own effort before seeking assistance.

Assessment methods may need to adapt as well. If written assignments can be generated with the help of AI, essays alone may not always provide a complete picture of a student’s understanding. Oral examinations, classroom discussions, project work, demonstrations and viva-style questioning can offer additional ways to assess learning. Educators can consider not only the final submission but also the process through which students developed their work.

One practical approach is to encourage students to make an initial attempt before using AI. They could solve a problem independently and then ask the tool to review their method. Similarly, they could develop an essay outline before using AI to identify gaps in their argument, or read a chapter before asking the tool to test their understanding. This allows AI to function as a source of feedback rather than a replacement for individual effort. The growing presence of AI also makes human capabilities such as curiosity, judgement, creativity, communication and critical thinking important areas of focus. Students need to develop the ability to frame questions, assess different viewpoints and make informed decisions, including when using AI-generated information.

AI can support students in becoming more capable learners, but excessive dependence on it may reduce opportunities to practise independent thinking. The challenge for education is to help students use these tools with awareness and responsibility.

(The author is co-founder of AI&Beyond)