Acentric Research is produced with AI tools under human review. It is affiliated with Acentric Capital, a private investment group whose members may hold positions in companies discussed. The authors are not financial professionals, and nothing here is investment advice or a recommendation to buy or sell any security. AI analysis can be wrong; check every claim against its cited source.
This Brief is an initial scan, not a full investigation. The method and scores behind it are in AI × Quantum in the Landscape Report.
Overview
This Brief covers the reverse of the other AI × Quantum topics. Instead of using artificial intelligence (AI) to run a quantum computer, quantum machine learning (QML) uses a quantum computer to run a learning method, such as a quantum version of a neural network. The hope is that for some tasks a quantum machine could learn from data faster or better than an ordinary computer.
A fair test of such a claim compares the quantum method with the best classical method on the same task, which is why “advantage” is the word the field uses. The sources opened give mostly aims and little demonstration. A European funding call opened in June 2026 and asks for methods that “can demonstrate clear advantages over classical approaches”. [1] A 2022 review in Nature Computational Science discusses opportunities for advantage and names the difficulty of training such models as a remaining challenge. [2] In the sources opened, no demonstration of an advantage over classical learning on a practical task was found, and a 2024 benchmarking preprint found the opposite on small tasks. [3] The sources opened date from 2022 to 2026; later developments were not searched.
Current state
What QML is. A 2022 review defines the field as sitting at the intersection of machine learning and quantum computing. Its abstract says QML has the potential of accelerating data analysis, especially for quantum data, with applications for quantum materials, biochemistry and high-energy physics. It reviews methods and applications, highlights differences between quantum and classical machine learning with a focus on quantum neural networks and quantum deep learning, and discusses opportunities for quantum advantage. [2] Quantum advantage means doing a useful task better than any classical method.
The European call. The European High Performance Computing Joint Undertaking (EuroHPC JU) describes its call as one that aims to strengthen Europe’s scientific and technological capabilities and advance research at the intersection of quantum computing and AI. The call opened on 2 June 2026 and closes on 28 January 2027. Its expected outcomes include development, validation and demonstration of QML methods, including quantum, quantum-inspired or hybrid algorithms, benchmarking and strategies for noise. Hybrid means a quantum processor working with a classical supercomputer. The call places particular emphasis on scalable solutions for large datasets and on quantum-native learning models that can demonstrate clear advantages over classical approaches. [1]
Federal work. DOE’s Genesis Mission opportunity lists a focus area on hybrid quantum-classical optimization using agentic AI in quantum chemistry and materials, and the July 2026 list of chosen projects includes one titled as using quantum machine learning and agentic AI for simulations of plasma dynamics. [4][5] The same focus area asks for AI-driven surrogate models, meaning fast learned stand-ins for slow simulations, to speed up quantum time-evolution and many-body simulations. [4] These are funding statements and titles, not results; the project title shows that quantum machine learning is one of the methods DOE has chosen for negotiation, but the list does not report any outcome.
Why the call stresses hybrid methods. DOE’s Genesis Mission opportunity, in its challenge on quantum algorithms, speaks of workflows that use the interplay of classical and quantum resources, with AI managing data flow and computation across platforms. [4] The EuroHPC call is likewise framed around combining quantum processors with classical high-performance computing to address computational bottlenecks while maintaining scalability. [1] The call is single-stage, with one deadline. [1]
The benchmark evidence. A 2024 preprint, with no journal reference found in its arXiv record, tested 12 popular quantum machine learning models on 6 classification tasks (160 datasets) in classical simulation. It reports that out-of-the-box classical models generally outperformed the quantum classifiers, and that removing entanglement from a quantum model often gave as good or better results. It cautions that the scales within reach today are small and that narratives influenced by commercialization make robust conclusions difficult. [3] This is one study of small tasks, and this Brief did not search for theoretical work on where an advantage might be proven.
Organizations. Apart from the funders above, this Brief did not search for companies or laboratories specializing in QML.
How far it has come. Early research, judging by the sources opened. In the sources opened, a review and a funding call report no demonstrated advantage, and a benchmarking preprint reports classical models generally doing better on small tasks. [3]
Opportunities
If a QML method showed a real advantage, the review’s abstract names quantum materials, biochemistry and high-energy physics as application areas. [2] The EuroHPC call is an explicit attempt to produce such demonstrations, with benchmarking as an expected outcome, which would show whether any advantage exists. [1] DOE’s focus area on hybrid algorithms is another route. [4]
Challenges
- Physics and evidence. The 2022 review names the trainability of QML models as a remaining challenge; the sources opened show no demonstrated advantage over classical methods. [2][1]
- Engineering. The call lists noise-aware strategies among its expected outcomes. [1]
Data sources
- CORDIS for the projects the EuroHPC call will award, once awarded. [6][1]
- DOE Genesis Mission opportunity and the list of chosen projects, for federal programs in this area. [4][5]
- Research literature, through Europe PMC or arXiv, for benchmarks that compare quantum and classical learning. [7]
- Benchmark and theory papers that compare quantum learning methods with the strongest classical methods; one preprint was opened for this Brief, and a deeper investigation would read more. [3]
- No registry or company list covers QML; the sources opened found no company in this niche.
Changelog
- 2026-11-18: First published.
Sources
- European High Performance Computing Joint Undertaking, “Quantum Machine Learning”, call HORIZON-JU-EUROHPC-2026-QML-07 (programme page; call status, dates and expected outcomes). eurohpc-ju.europa.eu (accessed 2026-11-18)
- M. Cerezo and others, “Challenges and opportunities in quantum machine learning”, Nature Computational Science 2, 567–576, published 2022-09-15 (a review). doi.org (abstract read through the Europe PMC API, record 38177473, accessed 2026-11-18; the full text is not open access and was not read)
- Joseph Bowles, Shahnawaz Ahmed and Maria Schuld, “Better than classical? The subtle art of benchmarking quantum machine learning models”, arXiv:2403.07059, 2024-03-11 (preprint; no journal reference was found in the arXiv record). arxiv.org (abstract read through the arXiv API, accessed 2026-11-18; the full text was not read)
- US Department of Energy, “The Genesis Mission: Transforming Science and Energy with AI”, Notice of Request for Application DE-FOA-0003612, issued 2026-03-17, Amendment 000003 of 2026-04-20; challenges 7 and 8. science.osti.gov (accessed 2026-11-18)
- US Department of Energy, “Energy Department announces first Genesis Mission projects to accelerate new era of AI-driven scientific discovery”, list of selections for award negotiations, posted 2026-07-22 (the list states that selection is not a commitment to issue an award or provide funding). energy.gov (accessed 2026-11-18)
- CORDIS, EU research results service, home page. cordis.europa.eu (listed as a data source, not opened for this Brief)
- Europe PMC, REST API (search and full-text endpoints), queried 2026-11-18. ebi.ac.uk