Matter & Energy

AI can't replace mathematical thinking, Polish study warns

Adobe Stock
Adobe Stock

Artificial intelligence and advanced mathematical software can solve increasingly complex equations, but engineers who rely on them without understanding the mathematics behind the answers risk making poor or even unsafe decisions, according to new research from Poland.

In a study published in the Canadian Journal of Science: Mathematics and Technology Education (doi: 10.1007/s42330-026-00488-z), Marek Małolepszy, PhD, of the Centre for Mathematics and Physics at Lodz University of Technology argues that engineering education should place greater emphasis on understanding mathematical concepts and applying them in real-world technical problems rather than simply teaching students how to perform calculations.

The findings come as AI-powered tools and mathematical software are becoming commonplace in engineering, capable of solving equations, calculating integrals, analysing data, generating graphs and running complex simulations. While these technologies reduce the need for manual calculations, Małolepszy argues they do not reduce the need for mathematical understanding.

Instead, he says, future engineers must know how to choose the right mathematical tools, understand the assumptions behind them and judge whether the results make physical and engineering sense.

The study highlights what Małolepszy describes as a "black box" approach to problem-solving, in which students enter a problem into a computer program and accept the result without understanding how it was obtained or whether it is appropriate for the engineering task at hand.

In engineering practice, technical problems rarely arrive in mathematical form. Instead, engineers must analyse issues such as structural stress, system stability, heat transfer, electrical signals or machine performance, translate them into mathematical models, select appropriate analytical methods and determine whether the results accurately describe the physical phenomenon.

To investigate how mathematics should be taught, Małolepszy interviewed 21 university lecturers teaching core engineering subjects. The interviews explored expectations of students' mathematical knowledge, the role of calculations, the growing use of computational software and the difficulties students encounter when applying mathematics outside mathematics classes.

The interviews revealed broad agreement that engineering students need a solid grounding in higher mathematics but do not need to memorise every formula, theorem or calculation technique. Instead, lecturers said students should understand mathematical concepts, recognise when particular methods should be used and apply them appropriately in technical contexts.

Basic calculation skills remain important, the study found, but they should support rather than overshadow the broader goal of defining problems, selecting suitable analytical tools, interpreting results and verifying their accuracy.

One of the biggest challenges identified by lecturers was students' difficulty transferring mathematical knowledge to subjects such as physics, mechanics, electronics and automation. Many students continue to treat mathematics as separate from engineering, even though equations, graphs, measurements, simulations and physical models often describe the same underlying phenomenon.

The study argues that mathematics and engineering should therefore be taught in closer coordination. According to Małolepszy, mathematics courses should retain a strong theoretical foundation while demonstrating how mathematical concepts are applied to real engineering problems. Digital tools, including AI, should become an integral part of that process—but only if students understand what the software is calculating, the assumptions behind the calculations and how to evaluate the results.

The findings suggest that as computers become increasingly capable of performing calculations, the most valuable skill for future engineers will not be carrying out mathematics by hand, but knowing what needs to be calculated, why it matters and whether the answer can be trusted before it is used to make engineering decisions. (PAP)

PAP - Science in Poland

kmp/ zan/

tr. RL

The PAP Foundation allows free reprinting of articles from the Nauka w Polsce portal provided that we are notified once a month by e-mail about the fact of using the portal and that the source of the article is indicated. On the websites and Internet portals, please provide the following address: Source: www.scienceinpoland.pl, while in journals – the annotation: Source: Nauka w Polsce - www.scienceinpoland.pl. In case of social networking websites, please provide only the title and the lead of our agency dispatch with the link directing to the article text on our web page, as it is on our Facebook profile.

More on this topic

  • 05.07.2015 PAP/Marcin Bielecki

    Scientists record rare gamma-ray burst just before lightning strike

  • Adobe Stock

    Researchers unveil programmable terahertz device that combines computing and communications for future 6G

Before adding a comment, please read the Terms and Conditions of the Science in Poland forum.