If the AI bubble bursts, what will be the difference in higher education and lifelong learning?

Abstract

As pointed out by Floridi (2024) and Youvan (2025), the current AI hype shows several characteristics that resemble earlier tech bubbles such as the Dot-Com Bubble and the Cryptocurrency Bubble. These two bubbles were also centred around a technology with a potential to revolutionise multiple sectors, and at the same time had a tendency to cause tunnel vision. The general pressure to meet high expectations, contradicted by reports on technological challenges and AI slop (Ansari, 2025), raise concerns about sustainability. If the AI bubble bursts there could be drastic consequences across tech industry, job market, and the general economy (Youvan, 2025). Less has been published about how a bursting AI bubble might have a strong impact on different educational contexts. This study has the aim of analysing and discussing how the fields of higher education and lifelong learning would be affected by an AI bubble burst. In the field of higher education most of the identified challenges related to GenAI will probably remain such as the need for rethinking assessment and course design (Perkins et al., 2025). There is still a need for concrete guidelines on how educators and student ethically might use generative AI (GenAI) in teaching and learning activities. Regarding lifelong learning, the need for an increase of lifelong learning has increased in modern society, creating new challenges for higher education to achieve lifelong learning. As lifelong learning includes learning throughout the entire life span, it can be seen as a way to strengthen individuals' opportunities to participate actively in society and to develop democratic and personal competencies. Here, the iterative improvement of GenAI tools could support flexibility individualisation, accessibility and individualisation for students. Further, lifelong learning is often emphasised in relation to the labour market's need for a well-educated and competent workforce (Jaldemark; 2021; OECD, 2021) of which competences in tools will be necessary in the future. If the AI bubble bursts, it may result in an interrupted development of GenAI tools. The existing toolbox works well to at least partly solve many types of traditional higher education assignments for students. The quality of speech synthesis, audio to text conversion, and tailored instructions are today much higher than just a few years ago. Students’ use of chatbots as artificial tutors and study buddies will not be ended by a burst AI bubble. At the same time, the iterative improvement of GenAI tools has reached a level of quality that could meet the high expectations on GenAI regarding individualisation and accessibility for students in higher education and therefore lifelong learning.