All illusions are wrong, but some are useful

While we agree with the general argument of Shiffrin and colleagues (2026) that scientists fall prey to the illusion of knowledge, we disagree with their argument that illusions present a fundamental problem for science. We argue that theory cannot exist without illusion, and therefore, our goal should not be to avoid illusions, but to create the best illusions possible. Specifically, we argue that many known illusions provide utility for both science and real-world applications, that attempting to dispel illusions is a fundamental aspect of science, and that illusions only become problematic when scientists become complacent and no longer seek to improve understanding. We conclude that researchers should continue to utilise the scientific method to understand what theories can, cannot, and should not predict, thus dispelling illusions through progressive research programs.