HSE Study Reveals Imbalance in the Generative AI Market
Researchers at HSE University analysed how effectively the global generative artificial intelligence market converts investment into real revenue, concluding that AI is currently developing faster than it is paying off. The results have been published in the journal Foresight and STI Governance.
In recent years, generative artificial intelligence (GenAI) has become one of the main areas of technological investment. Companies are pouring billions of dollars in chips, servers, and data-centre infrastructure, expecting rapid economic returns from large language models.
However, market expectations may be overstated. HSE University Academic Supervisor Yaroslav Kuzminov and Ekaterina Kruchinskaia, Associate Professor at the Faculty of Social Sciences and Senior Lecturer at the Department of Higher Mathematics, set out to assess the balance of the generative AI market and whether a gap exists between infrastructure investment and revenues from AI technologies.
The authors applied the DEA method—a model used to analyse the efficiency of complex economic systems based on multiple input and output parameters. In this case, the ‘input’ consisted of revenues of AI hardware manufacturers (chips, servers, semiconductors, and data-centre infrastructure), including companies such as AMD, Intel, and NVIDIA. The ‘output’ was the revenue of companies developing and monetising AI solutions, including Sony, OpenAI, Google DeepMind, Amazon, and Apple. In essence, the model simulates the AI market at both the input and the output stages, assuming that these players set the main agenda.
The analysis covers the period from 2016 to 2024. Importantly, the years themselves were treated as the units of analysis—although companies typically serve as units in this method. This decision was deliberate: the authors aimed to evaluate the efficiency of AI development overall in each specific year, rather than within individual companies. To test the robustness of the results, calculations were performed both in absolute terms and with adjustments for global GDP. This approach made it possible to assess the relative efficiency of the generative AI market across different years.
The analysis showed that the development of the GenAI market is nonlinear. As generative models emerged and underwent initial commercialisation between 2016 and 2021, efficiency increased. However, beginning in 2021 the trend changed: efficiency indicators declined despite a sharp rise in investment. After a short-term surge in 2023, efficiency again returned to the level recorded in 2022.
Ekaterina Kruchinskaia
‘From a purely methodological perspective, the results suggest that the AI solutions market is developing according to a catch-up model: revenues from software products do not yet compensate for the massive investment in hardware infrastructure. Increased demand for chips and computing power is driven by the development of large language models, but their commercial returns remain limited and do not offset the cost of hardware technologies or further investment in them,’ said Ekaterina Kruchinskaia.
According to the researchers, the current development model strengthens the position of hardware manufacturers but produces limited economic returns, as computing power becomes an end in itself. The market for AI solutions and applications capable of influencing social processes—for example, by increasing labour productivity—faces not only constraints such as the high cost of hardware and training runs, shortages of qualified personnel, and technological limits of the models, but also struggles to generate sufficient revenue, especially when compared with the scale of investment required.
Yaroslav Kuzminov
‘AI is indeed transforming not only the economy and companies’ business models but also everyday social life. This is evident in our daily lives. At the same time, its influence is actually spreading more slowly than it may appear and is less productive than many would like. Many people speak of a bubble in the AI market—a phenomenon not new to the global economy. It would be fair to say that such risks do exist. Our model opens the door to a more practical discussion in this direction. It is important to have not only analytical tools but also an applied plan, and a straightforward one at that. Without improving the efficiency of applied solutions, expanding their adoption and pursuing more balanced investment planning, further positive progress will be difficult,’ noted Yaroslav Kuzminov.
The authors emphasise that studies of this kind are important not only for the academic community but also for businesses, investors, and the development of balanced science and technology policy in the field of artificial intelligence.
See also:
Neural Network Maps as a Method for Constructing Mathematical Models
Scientists from HSE University–Nizhny Novgorod and the Institute of Physics Belgrade, Serbia, are jointly exploring the application of machine learning techniques and neural networks to the study of nonlinear dynamics. Natalya Stankevich, Leading Research Fellow at the Laboratory of Topological Methods in Dynamics of the Faculty of Informatics, Mathematics, and Computer Science at HSE University–Nizhny Novgorod, spoke to the HSE News Service about this international project.
HSE Scientists Develop Method to Compress Large Language Models Without Losing Quality
Researchers from the AI and Digital Science Institute at the HSE Faculty of Computer Science have developed a new compression method for large language models such as GPT and LLaMA that reduces their size by 25–36% without additional training or significant loss of accuracy. This is the first approach to use mathematical transformations—specifically, rotations of model weights—to make models more amenable to compression with structured matrices. The study results have been published in ACL Findings 2025. The code is available on GitHub.
Machine Learning Models Can Help Reduce Volatility and Boost Stock Market Returns
The use of machine learning models makes it possible to achieve greater accuracy in predicting risks in the Russian stock market compared to classical econometric approaches. The predictive power of these models increases by 23%, while the average investor’s return can reach up to 13% per annum. These conclusions were drawn by Nikita Lysenok from the Department of Financial Market Infrastructure at the HSE Faculty of Economic Sciences. The paper has been published in Fundamental and Applied Mathematics.
Pocket Money, Personal Interest, and Family Practices: What Shapes Students’ Economic Literacy?
University students' economic literacy depends not only on their field of study but also on their interest in economics, the learning environment, and family financial practices. For example, students who received pocket money irregularly tend to perform better on economic literacy tests than their peers who received financial support on a regular basis. These findings come from a study conducted by HSE University involving more than 1,100 students from five Russian universities. The findings have been published in Cakrawala Pendidikan.
‘Entering Robotics Now Means Growing with the Area’
Unmanned vehicles, courier robots, and smart speakers are rapidly becoming a part of our lives. In 2026, the HSE Faculty of Computer Science opens its new Bachelor’s Programme ‘Design of Intelligent Robotic Systems’ (DIRS). It will train specialists at the intersection of IT, artificial intelligence, and robotics. Academic Supervisor of DIRS Vadim Morgachev explains how studies are organised and why graduates of the programme ‘will definitely be accepted into the future.’
HSE Scientists Train Neural Network to 'Hear' Faults in Electric Motors
Researchers at the AI and Digital Science Institute of the HSE Faculty of Computer Science have developed a new method—the Signature-Guided Data Augmentation (SGDA) framework—that achieves 99% accuracy in motor fault detection and 86% accuracy in fault classification. The application of this approach can reduce industrial equipment repair costs, minimise downtime, and improve production safety. The study results have been published in Engineering Applications of Artificial Intelligence.
MIEM Tech Day at Pokrovka: Exploring HSE’s Engineering DNA Together
On May 26, 2026, the central atrium of the building at 11 Pokrovsky Bulvar will host the annual large-scale festival of engineering developments created by project teams from the HSE Tikhonov Moscow Institute of Electronics and Mathematics (HSE MIEM). The programme includes presentations of the best student technological projects, stands from partner companies and joint workshops, a lecture series featuring practising engineers, a round table on the development of engineering education, and presentations of MIEM master’s degree programmes.
The 'Second Shift' Is Not Why Women Avoid News
Women are more likely than men to avoid political and economic news, but the reasons for this behaviour are linked less to structural inequality or family-related stress than to personal attitudes and the emotional perception of news content. This conclusion was reached by HSE researchers after analysing data from a large-scale survey of more than 10,000 residents across 61 regions of Russia. The study findings have been published in Woman in Russian Society.
Resource Race and Green Transition: Three Unexpected Conclusions from Foresight Centre’s Research on Climate and Poverty
Beneath the surface of green energy—which most people associate with solar panels, electric vehicles, and reduced CO2 emissions—lies a complex web of geopolitical interests, international inequality, and resource constraints. Researchers from the Laboratory for Science and Technology Studies (LST) at the HSE ISSEK Foresight Centre have published a series of articles in leading international journals on hidden and overt conflicts surrounding critically important metals and minerals, as well as related processes in the energy sector.
Immersion in Second Language Environment Influences Bilinguals’ Perception of Emotions
Researchers at the Cognitive Health and Intelligence Centre at the HSE Institute for Cognitive Neuroscience have discovered how bilingual individuals process emotional words in their native (first) and non-native (second) languages. It was found that the link between word meaning and bodily sensations is weaker in a second language than in a first language. However, the more a person is immersed in a language environment, the smaller this difference becomes. The article has been published in Language, Cognition and Neuroscience.