Funded projects
\nBacked by £6 million over 9 months, these projects will test whether AI systems can plan and run scientific experiments in the real world.
Explore the research
\nThese projects reflect a striking diversity of technical approaches, from neurosymbolic models to vision-language systems for robotics. Projects span the UK, the US, and Europe, bringing together major platforms, leading universities, and emerging startups.
\nTogether, these teams are tackling a wide range of physical scientific challenges, including:
\nAs AI systems make hypothesis generation increasingly abundant, the bottleneck in science is shifting toward validation: the physical capacity to test ideas in the real world.
\nThese projects are structured as nine-month sprints designed to probe the limits of AI-driven discovery. Can AI Scientists recover when experiments fail? Can they identify interdisciplinary opportunities that human researchers might overlook? To answer these questions, each project will pursue two challenges: one the system is expected to solve, and one where it is likely to struggle.
","mediumLayout":false,"fullWidth":false},{"type":"creator-cards","items":[{"title":"Amina: Autonomous AI Scientist for Rapid Pathogen Diagnostic Design","description":"","teamLead":"Abhi Rajendran, AminoAnalytica","team":"Team: Adam Wu + Stefano Angioletti-Uberti, AminoAnalytica ","text":"COVID-19 showed that slow diagnostic development costs lives. For newly emerging pathogens, current diagnostic tests can take 6-12 months to design and validate, leaving public health systems reactive rather than prepared. This project is building an autonomous AI Scientist that designs rapid molecular diagnostics for emerging pathogens. By integrating biological design tools with automated reasoning and planning, the team aims to compress diagnostic development timelines from months to days. The project will evaluate whether an AI Scientist can reliably generate usable diagnostic designs under time pressure, helping shift pathogen response from crisis reaction to early prevention.
","label":"Active","cardId":"amina","mediaType":null,"modules":[]},{"title":"Wet-Lab-First AI Scientist","description":"","teamLead":"Katya Putintseva, Briefly Bio","team":"Team: Staffan Piledahl, Harry Rickerby + Hampus Ahlgren, Briefly Bio","text":"Most AI-driven laboratory systems optimise experiments within narrow, pre-defined workflows. This project takes a different approach: starting from flexible robotic execution as the core capability, and building an AI Scientist around it. This team has developed a system that can autonomously design and execute wet-lab experiments, with built-in guardrails that prevent unsafe or infeasible protocols. An end-to-end, LLM-based AI Scientist plans experiments, adapts to results, and directs laboratory robots in real time. The system will be applied to improving Agrobacterium-mediated plant transformation, the dominant gene delivery method in plant biotechnology and a critical bottleneck in plant engineering. The project will run alongside ARIA Creator, Syntato, who are designing synthetic plant chromosomes, providing a real-world test of autonomous experimental optimisation in a foundational biological process.
","label":"Active","cardId":"wet-lab-first","mediaType":null,"modules":[]},{"title":"Silico Habilis","description":"","teamLead":"Garik Petrosyan, Deep Origin","team":"Team: Ashot Papoyan, Khachik Smbatyan, Max Ratnikov, Tigran Abramyan + Garegin Papoian, Deep Origin; Mark Ofield + Michela Serena, Arctoris","text":"Many diseases remain poorly treated because their biology is complex and experimentally expensive to explore. This project deploys an AI Scientist to integrate biological simulations, literature analysis, and autonomous laboratory experimentation into a single discovery loop. Focusing on endometriosis, the system will analyse existing biological data and literature to identify underexplored therapeutic targets. It will then design candidate small molecules, which will be synthesised and experimentally tested in an autonomous laboratory. The AI Scientist will iteratively update its hypotheses based on experimental results, with human oversight limited to review and governance. The project tests whether tightly coupling simulation tools, automated labs, and AI reasoning can accelerate therapeutic discovery for complex diseases with minimal human intervention.
","label":"Active","cardId":"silico-habilis","mediaType":null,"modules":[]},{"title":"Automated Elucidation of Mechanisms Driving Age-Related Lysosome Failure","description":"","teamLead":"Michaela Hinks, Edison Scientific + Mathieu Bourdenx, University College London","team":"Team: Angela Yiu, Arvis Sulovari, Ludovico Mitchener, Benjamin Chang, David Colognori, Laurie McCoy + Evan McCabe, Edison Scientific","text":"Age-related decline in neuronal protein quality control increases the risk of neurodegenerative disease and may be driven by impaired lysosomal acidification. However, the molecular mechanisms behind this process remain poorly understood. This project uses Kosmos, an AI Scientist developed by Edison Scientific, to investigate the causes of lysosomal de-acidification during ageing. Kosmos will analyse publicly available datasets, generate mechanistic hypotheses, and design detailed experimental protocols to test them. Experiments will be executed in a semi-automated laboratory, after which Kosmos will analyse the results and produce an integrated report linking data, hypotheses, and conclusions. The project evaluates whether an AI Scientist can autonomously carry out mechanistic biological research – from data mining through experimental validation – in a complex ageing-related system.
","label":"Active","cardId":"automated-elucidation-mechanisms","mediaType":null,"modules":[]},{"title":"AI-Driven Cell-Free Energy Development and Optimisation","description":"","teamLead":"Scott Riggs, Find What Matters + Anton Jackson-Smith, b.next","team":"Team: Maxwell Shapiro, Joseph Lozier + Ash Eldritch, Find What Matters; Sharon Newman, Jonathan Calles + Anton Molina, b.next","text":"Engineering energy-producing biological systems is complex, requiring iterative design, testing, and optimisation across many interacting components. This project introduces IGOR (Iterative Guide and Orchestrated Research), a Bayesian, multi-agent AI co-scientist designed to operate closed-loop bioengineering workflows. IGOR will generate hypotheses, design experiments, interpret results, and author reports and publish data while interfacing with the b.next automated laboratory, open data, and compositional biology platform. The focus is on engineering and optimising metabolic energy modules in cell-free systems, where rapid iteration is possible without living cells. The project will test whether an AI Scientist can manage increasing biological complexity and demonstrate the feasibility of AI-driven design for synthetic cells and modular bioenergy systems.
","label":"Active","cardId":"ai-driven-cell-free-energy","mediaType":null,"modules":[]},{"title":"ThetaWorld","description":"","teamLead":"Otter Quarks","team":"","text":"This team is developing an AI scientist, ThetaWorld, that makes sense of unexplained data by building causal explanations. Their aim is to demonstrate the potential of ThetaWorld to push the frontier of science and world-modelling in relevant real world domains, in this case battery safety and degradation by applying their system to the development of causal explanations for real degradation mechanisms.
","label":"Active","cardId":"thetaworld","mediaType":null,"modules":[]},{"title":"Towards a Self-Reflective AI Scientist for Autonomous Sustainable Microbial Protein Biomanufacturing","description":"","teamLead":"Miao Guo, King’s College London","team":"Team: Yansha Deng + Chris Lorenz, King’s College London","text":"Sustainable protein production using microbial fermentation is central to the UK’s Net Zero and circular-economy goals, particularly when integrating waste-derived feedstocks. However, conventional bioprocess optimisation remains slow, inefficient, and labour-intensive. This project is building a self-reflective AI Scientist embedded within a new generation of self-driving laboratories. The system integrates AI-guided hypothesis generation, virtual-lab simulation, multi-fidelity optimisation, and molecular-robotic automation into a unified adaptive workflow. The AI Scientist will autonomously design, test, and refine hypotheses in real time, and will be evaluated through sustainable protein production in both batch and continuous-flow bioprocesses.
","label":"Active","cardId":"towards-self-reflective-ai-scientist","mediaType":null,"modules":[]},{"title":"Putting a (Better) Brain in the Mobile Robotic Scientist","description":"","teamLead":"Andrew I. Cooper + Gabriella Pizzuto, University of Liverpool","team":"Team: Francisco Munguia-Galeano, Hong Wang, Sam Harding, Xenofon Evangelopoulos + Abdoulatif Cissé, University of Liverpool","text":"Most closed-loop experimental systems optimise parameters without developing an explicit conceptual understanding of the underlying science. This limits scalability, interpretability, and performance in high-dimensional experimental spaces. This project aims to embed explicit reasoning and ontology building into a mobile robotic scientist. By integrating large language models into closed-loop experimentation, the system will construct and refine scientific concepts on the fly, rather than treating experiments as black-box optimisation problems. The goal is to create AI collaborators that can reason about experiments, explain their decisions, and guide exploration in a way that more closely mirrors human scientific thinking, bridging the gap between data-driven optimisation and conceptual understanding.
","label":"Active","cardId":"brain-mobile-robotic-scientist","mediaType":null,"modules":[]},{"title":"The Cancer AI Scientist Project","description":"","teamLead":"Lennard YW Lee, Gareth Bloomfield + Anthony Hsieh, University of Oxford","team":"","text":"Developing effective cancer vaccines can take 10–15 years, slowed by biological complexity, fragmented data, and disconnected experimental systems. This project investigates whether tightly integrated AI-driven infrastructure can accelerate progress. For the first time in cancer research, the project will unify AI models of tumour-immune recognition, laboratory automation, and sovereign AI supercomputing into a single discovery platform. Automated research pods will continuously generate, test, and refine cancer vaccine hypotheses at scale. By integrating modelling, experimentation, and compute into a closed loop, the project will assess whether AI Scientists can deliver a step change in the speed and efficiency of translating cancer immunology into patient-ready vaccine candidates.
","label":"Active","cardId":"cancer-ai-scientist-project","mediaType":null,"modules":[]},{"title":"MIND-MATTER: AI-Driven Discovery of Self-Learning Materials","description":"","teamLead":"Andrey Ustyuzhanin, Constructor Knowledge Labs","team":"Team: Wilfred van der Wiel, University of Twente; Roman Novikov, Constructor Tech + Mariia Snigireva, Constructor Knowledge Labs","text":"This project applies an existing AI Scientist workflow to the design of Reconfigurable Nonlinear Processing Units (RNPUs) – a class of physical computing hardware whose behaviour is governed by underlying charge-transport mechanisms. This project uses AI-guided hypothesis generation, experiment design, and data analysis to test which charge-transport mechanisms dominate different computational behaviours. Specifically, the AI Scientist will design and interpret experiments to discriminate between variable-range hopping (VRH) and space-charge-limited current (SCLC) as the mechanisms underlying static versus temporal computation in RNPUs. By iteratively proposing hypotheses, directing measurements, and updating models, the system will probe how physical charge transport gives rise to computational capability. The project addresses a key bottleneck in physical computing: the lack of transparent, testable links between material physics and computation.
","label":"Active","cardId":"mind-matter","mediaType":null,"modules":[]},{"title":"AI NanoScientist","description":"","teamLead":"Rafa Gómez-Bombarelli + Milad Abolhasani, Lila Sciences","team":"Team: Ben Steimle, Aniruddha Dey + Gustavo Malkomes, Lila Sciences","text":"This project will use Lila’s AI NanoScientist to autonomously address two challenges in colloidal nanoscience. First, it will test whether an AI Scientist can achieve reproducible, high-precision synthesis of colloidal nanoparticles within a constrained parameter space. Second, it will explore unmapped precursor chemistries to discover and optimise nanoparticles with tailored properties and improved stability. The AI NanoScientist operates as a closed-loop research system, autonomously generating hypotheses, designing experiments, executing them using robotic platforms, interpreting results in real time, and producing structured reports. Its AI agent supports sequential learning and mechanistically informed hypothesis formation, allowing experiments to adapt as evidence accumulates.
","label":"Active","cardId":"ai-nanoscientist","mediaType":null,"modules":[]},{"title":"Hermes: A Self-Improving AI Scientist to Discover and Refine DNA Delivery","description":"","teamLead":"Henry Lee, Cultivarium","team":"Team: Nili Ostrov, James Knight, Kerrin Mendler, Stephanie Brumwell, Melanie Abrams + JooHee Choi, Cultivarium","text":"DNA delivery remains a fundamental bottleneck in biology, limiting our ability to study and engineer many organisms and forcing progress to rely on slow, bespoke, and labour-intensive protocols. This project will develop Hermes, a self-improving AI Scientist that autonomously designs, executes, and iteratively refines experiments to achieve DNA delivery in genetically intractable organisms. Hermes will combine hypothesis generation, experimental planning, robotic execution, and data analysis in a closed loop, learning systematically from both successes and failures. The project aims to move biological experimentation away from artisanal protocol tuning toward scalable, learning-driven systems. By embedding scientific judgment into software and automated laboratory workflows, the team will test whether an AI Scientist can accelerate discovery through structured iteration and evidence-driven experimentation.
","label":"Active","cardId":"hermes","mediaType":null,"modules":[]}]},{"type":"spacer-comp","cssSizeClass":"medium","height":0},{"type":"news-insights","title":"","sliderTitle":"","sliderLink":null,"featured":{"title":"The UK government is backing AI that can run its own lab experiments","text":"MIT Technology Review
\nBy funding a range of projects for a short amount of time, the agency is taking the temperature at the cutting edge to determine how the way science is done is changing, and how fast. What it learns will become the baseline for funding future large-scale projects.
","image":{"src":{"mobile":"/media/utsbptpf/angie-in-discovery.jpeg?width=300&height=375&format=webp&v=1db52096c096560","tablet":"/media/utsbptpf/angie-in-discovery.jpeg?width=440&height=550&format=webp&v=1db52096c096560","desktop":"/media/utsbptpf/angie-in-discovery.jpeg?width=600&height=750&format=webp&v=1db52096c096560"},"alt":"Angie, a Programme Director, in a lab coat and goggles in a lab during discovery. There is also a person in the background speaking and gesturing with their hands. ","title":""},"link":{"target":"_blank","isActive":true,"text":"Read more ","href":"https://www.technologyreview.com/2026/01/20/1131462/the-uk-government-is-backing-ai-scientists-that-can-run-their-own-experiments/","title":"Read the full article at MIT Tech Review"}},"items":[]},{"type":"spacer-comp","cssSizeClass":"large","height":0}],"scriptsAtTop":" \n\n\n\n\n\n\n\n\n\n\n","scriptsAtBottom":"","id":3321,"languageCode":"en","pageName":"Funded projects","scrollSmoothness":0.0,"gtmCode":"GTM-T8Z5F55","topNav":{"logo":{"text":"ARIA - Home","href":"/media/hsxblxrf/logo-aria-black.svg"},"nodes":[{"text":"About us","href":null,"target":null,"id":null,"nodes":[{"text":"About us","href":"/about-aria","target":null,"id":"2207","nodes":[],"noOpener":false,"noReferrer":false},{"text":"Our team","href":"/about-aria/our-team","target":null,"id":"2159","nodes":[],"noOpener":false,"noReferrer":false},{"text":"Programme Directors","href":"/about-aria/our-team/programme-directors","target":null,"id":"2831","nodes":[],"noOpener":false,"noReferrer":false},{"text":"Reporting and policies","href":"/about-aria/reporting-and-policies","target":null,"id":"2160","nodes":[],"noOpener":false,"noReferrer":false}],"noOpener":true,"noReferrer":false},{"text":"What we do","href":null,"target":null,"id":null,"nodes":[{"text":"How we work","href":"/how-we-work","target":null,"id":"2817","nodes":[],"noOpener":false,"noReferrer":false},{"text":"Live opportunity spaces","href":"/opportunity-spaces","target":null,"id":"2215","nodes":[],"noOpener":false,"noReferrer":false},{"text":"Emerging areas","href":"/emerging-areas","target":null,"id":"2963","nodes":[],"noOpener":false,"noReferrer":false},{"text":"Activation Partners","href":"/activation-partners","target":null,"id":"2174","nodes":[],"noOpener":false,"noReferrer":false}],"noOpener":true,"noReferrer":false},{"text":"Funding","href":null,"target":null,"id":null,"nodes":[{"text":"Funding opportunities","href":"/funding-opportunities","target":null,"id":"2164","nodes":[],"noOpener":false,"noReferrer":false},{"text":"Applicant guidance","href":"/funding-opportunities/applicant-guidance","target":null,"id":"2162","nodes":[],"noOpener":false,"noReferrer":false},{"text":"Being funded by ARIA","href":"/funding-opportunities/the-creator-experience","target":null,"id":"2433","nodes":[],"noOpener":false,"noReferrer":false},{"text":"FAQs","href":"/funding-opportunities/faqs","target":null,"id":"2649","nodes":[],"noOpener":false,"noReferrer":false}],"noOpener":true,"noReferrer":false},{"text":"Careers","href":"https://aria-jobs.teamtailor.com/","target":"_blank","id":null,"nodes":[],"noOpener":false,"noReferrer":false}],"headerLinks":[{"target":"","isActive":false,"text":"Become a Programme Director","href":"/about-aria/our-team/programme-directors/cohort-3","title":""},{"isActive":false,"text":"Opportunity spaces","href":"/opportunity-spaces","title":""}],"menuLinks":[{"isActive":false,"text":"Contact us","href":"/contact","title":""},{"isActive":false,"text":"Terms of use","href":"/terms-and-conditions","title":""}],"activeItem":"3321","menuText":"Menu","menuTextClose":"Close menu","search":{"isActive":false,"text":"Search","href":"/search","title":""},"animationOffText":"Motion: Off","animationOnText":"Motion: On"},"footer":{"header":"Help us drive momentum ","text":"ARIA is a non-departmental public body, sponsored by the Department for Science, Innovation + Technology.","copyright":"Copyright 2026","headerLinks":[{"isActive":false,"text":"Contact","href":"/contact","title":""},{"target":"_blank","isActive":true,"text":"Careers","href":"https://aria-jobs.teamtailor.com/","title":""}],"socialHeader":"Follow our progress","footerLinks":[{"isActive":false,"text":"Privacy Policy","href":"/privacy-policy","title":""},{"isActive":false,"text":"Terms and Conditions","href":"/terms-and-conditions","title":""},{"isActive":false,"text":"Cookie Policy","href":"/cookie-policy","title":""},{"isActive":false,"text":"Accessibility Statement","href":"/accessibility-statement","title":""},{"isActive":false,"text":"Accessibility Support","href":"/funding-opportunities/accessibility-support","title":""},{"target":"_blank","isActive":true,"text":"Careers","href":"https://aria-jobs.teamtailor.com/","title":""},{"isActive":false,"text":"Supply to us","href":"/supply-to-us","title":""}],"socialItems":[{"link":{"target":"_blank","isActive":true,"text":"ARIA Research Twitter","href":"https://x.com/aria_research","title":""},"type":"x"},{"link":{"target":"_blank","isActive":true,"text":"LinkedIn","href":"https://www.linkedin.com/company/advanced-research-and-invention-agency","title":""},"type":"Linkedin"},{"link":{"target":"_blank","isActive":true,"text":"ARIA Research Substack","href":"https://ariaresearch.substack.com/","title":""},"type":"SubStack"}],"formText":"","formSubmit":"Sign up","formId":"","recaptchaSiteKey":"6LcuyY4bAAAAAE2OzRFfA7vDWyohfZXyl2uTuHLa","verificationToken":"","redirectUrl":null,"submitMessage":"","formTheme":""},"seo":{"title":"Funded projects","description":"","keywords":"","includeInSitemapXml":true,"canonicalTag":""},"sharing":{"og":[{"property":"og:title","content":"Funded projects"},{"property":"og:image","content":"https://aria.org.uk"},{"property":"og:url","content":"https://aria.org.uk/ai-scientist/funded-projects"}],"twitter":[{"name":"twitter:card","content":"summary_large_image"},{"name":"twitter:site","content":"@ARIA_research"},{"name":"twitter:title","content":"Funded projects"},{"name":"twitter:image","content":"https://aria.org.uk"},{"name":"twitter:url","content":"https://aria.org.uk/ai-scientist/funded-projects"}]},"pageType":"contentPage"}