Nature: 'Designing physics experiments with artificial intelligence' (Krenn group) on AI-found setups that beat human designs
Published in Nature around Sept 2–3, 2026, Klimesch, Arlt, Ruiz-Gonzalez et al. (Mario Krenn's group, Tübingen, with TU Wien and Vienna) describe how search and optimization algorithms explore huge spaces of lab components to propose experimental setups that give more precise results or new measurement capabilities than human designs, across quantum optics, electron microscopy, fusion, particle detectors and gravitational-wave detectors.
Key facts
- Citation (Tübingen AI Center): Klimesch, J., Arlt, S., Ruiz-Gonzalez, C. et al. 'Designing physics experiments with artificial intelligence', Nature 657, 47–58 (2026), doi 10.1038/s41586-026-10898-6
- University of Vienna news dated Sept 3, 2026; Tübingen AI Center news dated Sept 2, 2026
- Method: optimization over mathematical models of available components (not a chatbot/LLM); humans set goals and constraints
- Application areas: quantum experiments, electron microscopy with entanglement, fusion reactors, particle detectors, gravitational-wave detector sensitivity
- Krenn: 'Human work is simply shifting to a higher level'; some AI-found designs are provably better but lack an intuitive explanation
Science result
- Field
- physics / experimental design
- Problem
- Automated design of physics experiments
- Result
- Overview and results showing AI-designed experimental layouts that outperform or extend human designs in several fields of physics.
- AI system
- Krenn-group optimization algorithms
- Human role
- Human-led: researchers define objectives and verify designs; AI searches the configuration space
- Verification
- Peer-reviewed in Nature
- Status
- confirmed
What happened
Krenn began this line of work as a student in Vienna, when an algorithm found a quantum-optics setup his group could not design by hand. The Nature paper extends the approach across physics; TU Wien's Philipp Haslinger said the AI proposed microscope designs "a human would probably never have come up with".
Why it matters
It shows AI in science moving from analysing data to designing the instruments and experiments themselves. Confidence is medium on details because the Nature text is paywalled; the press releases give no quantitative improvement figures.
Changelog
- 2026-09-30: created (resolves the Krenn part of the leads.md 'missed pre-window items' line)
Sources (4)
- paperNature: Designing physics experiments with artificial intelligence
- officialUniversity of Vienna: Artificial intelligence suggests new physics experiments
- officialTübingen AI Center: AI could help scientists design experiments humans would never think of
- pressPhys.org: AI suggests new physics experiments that could outperform human-designed setups
id: 2026-09-02-krenn-nature-designing-physics-experiments-ai · updated 2026-09-30 · open in the interactive timeline