TUM Campus Heilbronn strengthens its expertise in chip design

Modern AI applications rely not only on powerful software, but increasingly on specialized hardware as well. Expertise in chip design is therefore a key foundation for energy-efficient AI applications, embedded systems, industrial automation, and connected machines and products. With this new specialization, TUM is strengthening its expertise at the intersection of information engineering, microelectronics, and AI, and educating specialists for a field of significant technological and economic importance. At the same time, TUM is linking its research to the Heilbronn region's innovation and industrial ecosystem and creating connections among academia, innovation centers, and technology-oriented companies.

Can cooking help prevent colorectal cancer?

On a stovepot, onion skins, parsley stems and other vegetable scraps that would normally end up in the compost are gently simmering. Next to them, plantains are cooking. A young man chops walnuts while, across the counter, pitted dates are spread out on a baking sheet. Crumbled tofu mixed with peanut butter sizzles in a frying pan. Between cutting boards and pots, laughter fills the room. Around 20 participants have gathered on this Saturday in the teaching kitchen at the Institute of Clinical Nutrition Medicine on TUM’s Olympiapark campus. Together with the student team behind Science Snack, they prepare plant-based dishes ranging from dips to dessert. Every ingredient has been chosen with care. The recipes are all built around a question many people rarely consider in everyday life: What role does nutrition play in gut health? “Food brings people together, and when you’re chopping, cooking and tasting side by side, it becomes much easier to talk about topics like gut health and colorectal cancer prevention,” says chef Viet Ha Dietrich Nguyen. “It doesn’t feel like a lecture – it feels like a normal conversation.”

Launching a start-up through a master’s thesis

“Creating value out of nothing – that idea fascinated me immensely.” While studying informatics, Nils Börner realized he wanted to found a start-up. But with what product? “For a long time, I didn’t have an idea. As my studies progressed, I learned that you can also systematically search for a market niche. But if the Entrepreneurial Masterclass hadn’t been there as a starting point, things would probably have turned out very differently.” Guiding entrepreneurial-minded students and putting research findings from master’s theses into practice is the idea behind the TUM Entrepreneurial Masterclass, which is unique in Germany. “We want to leverage the untapped potential that lies dormant in master’s theses and that also stemmed from the fact that academic studies and the entrepreneurship ecosystem sometimes didn’t align,” says program manager Tim Bernhard. Over the first six years of the program, more than 500 participants have contributed to the creation of a start-up – either their own or, directly or indirectly, other companies. That’s because there are three different tracks in the Masterclass. Nils Börner chose the Startup Creation Track. “The application deadline really sped up the search for a business idea,” says Börner with a smile. He focused on a problem in freight forwarding: unnecessary routes that forklifts take when handling goods. In his master’s thesis, Börner investigated how this problem could be solved using mathematical optimization.  “The TUM Entrepreneurial Masterclass isn’t about writing a business plan and submitting it as a thesis,” emphasizes Prof. Anne Tryba, one of the academic directors. “The academic standards are as high as for all TUM theses. Students tackle a problem whose relevance extends beyond their own start-up project.”

TUM develops intelligent 3D twins of crime scenes

The Bavarian State Criminal Police Office already uses 3D models of crime scenes today. Hundreds or even thousands of photographs are combined to create digital reconstructions that can support investigations long after evidence has been collected. In their joint research project, however, TUM and the BLKA are taking this concept a step further: the virtual crime scene should not only be viewed, but also queried. “I am looking for a red jacket,” “How many knives are in the room?” or “Show me all sharp objects” – in the future, specialized AI software will be able to answer questions like these directly. To make this possible, the researchers are developing methods that automatically identify, classify, and catalogue objects within a digital crime scene. Objects of interest are highlighted in the model and can immediately be incorporated into investigative work.

Testing materials for fusion power plants

Inside a fusion reactor, materials are simultaneously exposed to extreme temperatures, mechanical stress, and intense particle radiation. This alters the crystal lattice, creates defects and pores, and can cause material failure. Affected materials include ferritic steels such as EUROFER97, tungsten and tungsten alloys for the reactor inner wall, as well as copper alloys for cooling components. Under neutron bombardment, rhenium also forms in tungsten - a change that affects the material properties. This is where FUMA comes in, the “Competence Center for the Optimization of Fusion Materials.” It brings together one positron research group and three neutron research groups from FRM II, as well as two IPP groups whose methods complement one another. One group is investigating additively manufactured components and welded joints, while another is analyzing pores and internal stresses under industry-relevant conditions. “With our positron spectroscopy at FRM II, we want to understand material damage at the atomic level,” explains project coordinator Prof. Christoph Hugenschmidt of TUM. “Using the antiparticles of electrons, we can trace how the smallest defects develop into larger structures that ultimately determine the service life of a material.” Prof. Christian Pfleiderer, Scientific Director at FRM II, adds: “The neutron methods used at FRM II provide unique insights into the structure of fusion materials, from the atomic scale to the millimeter scale.” Hugenschmidt explains why Garching offers ideal conditions for the project: “Here on site, IPP contributes outstanding expertise on materials and components for a future fusion reactor.” In addition, the site offers unique experimental capabilities: heat loads of up to 20 megawatts per square meter, as well as targeted material damage caused by neutron irradiation, can be studied under realistic conditions. The long-term goal is to be able to reliably predict the service life of materials in a fusion reactor. The groups are jointly developing a measurement program: IPP defines the research priorities for materials-related questions and prepares samples, while the analysis groups at FRM II continue to improve their neutron- and positron-based instruments and investigate the damage mechanisms.

Medical AI – a look ten years into the future

The digital platform is now available at https://gesund-mit-ki.de/ (German only). It presents concrete, everyday scenarios across five stages of life. How might intelligent wearables support Clara, 34, and her unborn child throughout pregnancy in 2036? How could AI help diagnose the rare condition of 16-year-old Lena? How might it save the life of Sebastian, 55, after a stroke? And how will assistive robots support older people in need of care ten years from now? Each story is based on the expertise of researchers at the Technical University of Munich (TUM), Ludwig Maximilian University (LMU), and Helmholtz Munich, and reflects the current state of research. In addition to these patient stories, the researchers also address broader questions such as: “Will people with serious illnesses increasingly be filtered out before birth?” and “How will drugs be developed in the years ahead?” The platform’s content will be continuously expanded in the coming weeks.

How fair does AI seem in job interviews?

Companies are increasingly using artificial intelligence in their hiring processes. It’s not just CVs that are evaluated automatically. AI tools can also conduct job interviews – usually in the form of avatars, which are animated characters – and make hiring decisions. An important reason for this, aside from saving time, is that AI is said to be less biased than humans. “What has been largely overlooked so far is that we all unconsciously react to the avatars’ appearance, even when we know we’re talking to a machine. That’s why a conversation with artificial intelligence becomes a social interaction as soon as it behaves like a human,” says Enkelejda Kasneci, professor of Human-Centered Technologies for Learning at the Technical University of Munich (TUM). A team of researchers from TUM and Lund University therefore investigated how applicants perceive AI decisions in job interviews depending on the avatar’s appearance. For the study, approximately 220 participants from Germany, the United Kingdom and the United States conducted a short simulated job interview for a fictional position in a company’s customer support department. They were interviewed by a photorealistic avatar capable of reacting to their answers and asking follow-up questions in a human-like manner. The research team had programmed four variants to embody the AI. The avatars were either female or male and had either dark or light skin. The participants’ eye movements were monitored using eye-tracking technology. After the interview, they completed a questionnaire.

Prof. Peus and Prof. Winkelmann re-elected as Vice Presidents

Management researcher Claudia Peus has been responsible for the area of Talent Management and Diversity since 2017. This also includes topics such as equal opportunity and continuing education. In addition, she heads the TUM Institute for LifeLong Learning (IL3). Physician and neurogeneticist Juliane Winkelmann has been responsible for the area of International Alliances and Alumni since 2017 and thus, in a sense, serves as TUM’s foreign minister. Seats on the Executive Board are assigned for three-year terms. In total, the Executive Board comprises 14 members. The body leads TUM and is responsible for its strategic higher education policy goals and the dynamic development of the university of excellence.

Better prediction of volcanic eruptions

The more forcefully lava rises from the Earth's interior toward the surface, the more gases are released. Carbon and sulfur compounds are good indicators of current activity in a volcanic field. In particular, the ratio of carbon dioxide to sulfur dioxide provides insight into what is happening beneath the Earth’s surface. Until now, these measurements had to be taken from ground level. The drawback is that the gases measured there do not originate solely from volcanic activity but are also emitted by surrounding vegetation and soil. To minimize these background signals, drones are now being used to fly above the gas clouds. “This is more precise and safer,” says Prof. Achim Lilienthal, deputy director of the TUM MIRMI Robotics Institute and head of the Chair of Perception for Intelligent Systems at the TUM School of Computation, Information, and Technology. Autonomous measurement of gases over a volcanic field As part of the DFG research project “Measurement Technology on Flying Platforms,” his research team has now developed a system capable of determining gas concentrations over an active volcanic field with high precision. TUM researcher Marius Schaab has now deployed the system autonomously for the first time on the Aeolian island of Vulcano, off the coast of Sicily. He has mounted a laser on a small cart that automatically locates an airborne drone and aligns itself with a reflector on the drone. The reflected beam is slightly weakened as it passes through the gas cloud because it is absorbed by the gas being measured — in this case, carbon dioxide.

The rules of reactions

Durst and TUM Venture Labs cooperate on robotics and AI

Durst develops intelligent systems for industrial digital printing and production. Headquartered in Brixen, this family-owned company is a pioneer in the transformation toward connected, data-driven production environments through the use of digitalization, automation and system integration. Each of the twelve TUM Venture Labs specializes in a key technology. In these fields, they offer start-up teams direct access to cutting-edge research, specialized technical infrastructure, tailored training programs, market expertise and a global network of industry contacts and investors. The Venture Lab Robotics / AI brings together expertise in robotics, artificial intelligence, embedded systems and industrial automation. At the center of the collaboration are regular expert exchange, industry challenges with start-up teams developing solutions, and the joint development of entrepreneurship and innovation formats. Durst contributes real production-related challenges where automation, robotics, AI-based process control and physical workflows converge. The resulting solutions are intended not only for the printing industry, but also for adjacent and cross-industry applications.

“AI is just one small part of the wider world of football analytics”

You previously outlined the potential applications of AI-powered game analysis for the 2024 European Championship. Have there been any new developments over the past two years? Yes, significant progress has been made in this field. Personally, I find skeletal data, i.e. data on each individual limb, particularly interesting. Thanks to advances in camera technology and computer vision models, this data can now be captured with good accuracy. What is this data used for? The collection of skeletal data was initially introduced mainly for video assistant refereeing, such as the semi-automatically detection of offside positions, and for creating animations from a player’s perspective to make match coverage more engaging. However, it is also highly valuable for match analysis, as positions such as those of shoulders, ankles, ears and the nose, as well as knee angles, can be used to answer performance-related questions. How do nose movements help with match analysis? By knowing the positions of the ears and the nose, we can estimate where a player is looking. From this, you can infer “situational awareness”, such as which teammates and opponents a player actually saw when passing the ball. If the AI identifies that a player is focusing too focused on the ball and not scanning their surroundings enough, this can be addressed in training. Can other insights be derived from this? At this year’s TUM Football Analytics Hackathon, we asked students to extract insights from skeletal data. This led to some very interesting projects. For instance, changes in running parameters such as stride length and leg extension were used to estimate player fatigue.

TUM ranked 8th in training digital leaders

Which universities best prepare their students to work with digital technologies? Which graduates contribute to the innovative strength of the digital economy? Who is founding tech start-ups or driving the transformation of large companies? To find out, the consulting firm Emerging identified more than 200,000 digital professionals worldwide and tracked their university degrees. Ranked 8th, TUM has made the top 10 of the Digital Leaders in Higher Education rankings for the third consecutive year as the only university in the European Union. The ranking is based on several categories. In Computer Science, it identifies graduates who develop digital technologies. Here, TUM ranks 2nd worldwide. The Entrepreneurship & Innovation category includes founders who bring new digital products to market with TUM ranking 5th. In Digital Transformation, the ranking counts graduates who are shaping organizational change in decision-making positions. In this category, TUM ranks 14th. Other rankings also show that TUM prepares its students exceptionally well for the professional world. In the Global University Employability Ranking, which surveyed approximately 12,000 managers in more than 30 countries, TUM ranks 13th. In the CHE University Ranking, students consistently give very high ratings for study conditions and course offerings.

First permanent building in Europe by star architect and TUM Professor Francis Kéré

The building was financed, constructed and donated to TUM by entrepreneur and TUM Honorary Senator Ingeborg Pohl. As the client, she commissioned Prof. Francis Kéré for the visionary design and his professorial colleague Hermann Kaufmann, who made the design feasible in the first place, for the execution planning. The result is a timber structure with five stories and an iconically folded louvered façade made of weathering steel. A total area of around 700 square meters is available for the children. The children are grouped by age; each age group occupies its own floor, and the middle and upper levels hold communal areas for play, sports, and meals, including a multipurpose sports room. At the very top, a partially covered rooftop terrace called the Himmelswiese (“sky meadow”) gives the children a sheltered place in the densely built urban environment to run, feel the wind and the sun, and look out across the city. At the heart of the design is the vertical playground. Slides connect the floors, making movement to a level below an invitation to play. The playground is also an acoustic buffer that shields the quieter rooms behind it from street noise.

How autonomous systems learn to understand their environment

It all began at a climbing gym. That’s where Daniel Cremers, professor for Computer Vision and Artificial Intelligence at the Technical University of Munich (TUM), met his future co-founder, Lukas Köstler. Shortly thereafter, Lukas Köstler began conducting research as a doctoral student in Professor Cremers’ department; Simon Klenk also joined the team there. As a result of this collaboration, the three decided to continue their research in their own company. With SE3 Labs, the team aims to give machines a spatial understanding of their environment so that they can act independently within it - similar to how generative AI handles language. Their startup receives support from UnternehmerTUM and TUM Venture Labs.

Special glove helps people with paralyzed hands grasp objects

The “soft-hand exoskeleton” consists of a fabric glove developed by the researchers, with air cushions attached to its outer surface. The air cushions are inflated through a total of 13 tubes, providing targeted support for the hand movements needed to hold a plate or grasp a glass, fork or spoon. The air-filled cushions allow each finger to be bent and straightened individually while also rotating the wrist, enabling objects to be held securely in the hand. To determine when a person intends to grasp an object, the researchers measure muscle activity in the forearm. Sensors attached to the forearm capture electrical signals, which are analyzed using machine learning to reliably determine the intended movement. “To prevent objects from being dropped accidentally, we use additional motion sensors to detect transport movements and keep the exoskeleton’s grip securely closed throughout the movement,” says researcher Nicolas Berberich. A soft-hand exoskeleton that anyone can afford

Programmable molecular machines are getting closer

The switch is based on DNA origami, where DNA strands are folded into precisely defined, nanoscale components. Using this technique, the team created a DNA switch with two stable positions. A short electric pulse is sufficient to move the structure from one position to the other within milliseconds. After that, it remains in its new position without any further energy input. This is an important step for the development of molecular machines. Such systems must not only be switched in a controlled manner, but also operate reliably over extended periods. This is exactly what the new switch demonstrates: In the experiments, individual devices remained stable for hours, withstanding more than 200,000 switching cycles and, in an additional setup, still showing robust switching behavior even after around one million actuations.

Five ERC grants awarded to TUM researchers

To date, researchers at TUM have secured a total of 270 of these highly competitive ERC grants. They are awarded annually in various categories. ERC Proof of Concept grants are given to researchers who want to investigate whether their ERC research projects can lead to commercially viable innovations.

Study reveals privacy risks in medical AI

Researchers at the Technical University of Munich (TUM), Imperial College London, and the Hasso Plattner Institute (HPI) demonstrate that earlier calculations of AI model security are misleading. Attacks that aim to determine whether an individual’s data was used to train a model are known as membership inference attacks (MIAs). Until now, common medical AI models were considered largely secure against MIAs. “Unfortunately, previous risk assessments have only ever measured the average risk across all patients. We examined the risk at the level of individual patients for the first time – and it paints a very different picture,” says TUM researcher Moritz Knolle, first author of the study. While the attacks were unsuccessful for a large proportion of the datasets, some patients could be linked to the models with near‑100 percent certainty. “This is not a tolerable risk. Health data is highly sensitive,” says Daniel Rückert, Professor of Artificial Intelligence in Healthcare and Medicine at TUM and, together with Professor Georg Kaissis (HPI), senior author of the study.

“AI is fundamentally changing the way we work”

Digitalization, new ways of working, increased complexity—our world is changing dramatically. How can I, as an employee, keep up? To me, “staying on top of things” means pursuing targeted continuing education whenever new digital tools and relevant developments emerge in my field of work. Artificial intelligence, in particular, is fundamentally changing the way we work. You can’t just “stock up” on knowledge—instead, you should regularly schedule small learning sessions and always try out new things right away. And if you don’t feel confident doing this on your own, you can ask your colleagues. After all, constantly learning new things is a task that affects us all, isn’t it? Many questions and challenges are similar, and it’s much more fun to tackle them together! The Online Learning Days 2026 provide a great framework for this—compact, practical, and easily applicable short workshops in which experts demonstrate the new digital tools available at TUM. What’s on offer at the Learning Days? The Learning Days 2026 bring together a wide range of offerings related to digital work and artificial intelligence. Over the course of three days, employees can learn about and try out specific tools in eleven workshops. The spectrum ranges from the TUM Trust Center to the TUM Wiki and Microsoft Copilot. Will participants be able to get hands-on experience in the workshops? The focus is clearly on trying things out, understanding them, and applying them right away. In addition to a “guided tour” of each tool, the 75-minute sessions also leave time for specific questions. Who can participate? The Learning Days are open to all TUM employees—regardless of whether they’ve had some exposure to digital tools or already have extensive experience and want to expand their knowledge in specific areas. So, the bottom line is: No prior knowledge required! The events are particularly geared toward employees in administration, research management, and other research-supporting areas who want to make their daily work more digital. How can I register? You can register up until the day before the workshop at: www.tum.de/learningdays. I’d be delighted if we could surpass last year’s registration numbers.

Presidential Award for space start-up

In 2025, researchers, students and alumni of the Technical University of Munich (TUM) founded 115 start-ups, a new record. The enthusiasm for starting businesses remains strong. The number of initial consultations at the TUM Start-up Consultancy also reached an all-time high of around 500. Last year, TUM received 32 approvals for EXIST start-up scholarships – by far the highest number nationwide. Through these grants, the Federal Ministry for Economic Affairs supports spin-offs from universities when the universities themselves support the teams.  More than 40 spin-offs presented themselves to the approximately 1,500 visitors at TUM Entrepreneurship Day. Key players in Munich’s innovation ecosystem discussed topics including quantum technology companies. TUM researchers presented a new study on which universities are shaping the start-up landscape. The conference “From Lab to Launch – Unleashing Academic Start-ups” was broadcast live. At the event, held at the TUM Heilbronn Campus, international guests, including researchers from the universities of Oxford, Cambridge, Stanford and Berkeley, as well as Imperial College London, discussed new approaches for spin-offs. A particularly promising spin-off was awarded the TUM Presidential Entrepreneurship Award. The award recognizes business ideas that are largely based on research findings, have high growth potential and have already secured initial funding. It comes with a prize of 10,000 euros, donated by the TUM Association of Alumni and Friends.

Start-ups create 25,000 jobs in 10 years

Last year, the study “Entrepreneurial Impact of Academic Institutions” provided the first comprehensive overview of where start-up founders in Germany, Austria and Switzerland had studied and conducted research. Now, the research team from the Technical University of Munich (TUM), ETH Zurich and the University of Innsbruck has further analyzed the data on the approximately 51,000 start-ups founded in the three countries between 2014 and 2024. The results presented today at TUM Entrepreneurship Day provide an important foundation for entrepreneurship support and higher education policy and reveal which universities are shaping the start-up landscape.

Lifelong learning and people development

How Robots Become Coworkers

When robots and humans work together in a factory, they are usually kept clearly separated—in part for safety reasons. The robot takes on a task, completes it, and hands it over to workers, who carry out the next step in the process. “This can be physically demanding, for example, during quality inspections of components that have to be repeatedly lifted and set down,” says researcher Federico Masiero from the Chair of Intelligent Bio-Robotic Systems at the TUM School of Computation, Information and Technology. Developed at the Munich Institute of Robotics and Machine Intelligence (TUM MIRMI), led by Prof. Lorenzo Masia, the so-called WearaCob (short for “wearable” and “collaborative”) is designed to make lifting and carrying easier for people with the help of an upper-body exoskeleton, enabling assistance from a single-arm, collaborative robot, or “cobot”.  Exoskeleton: Reduces upper-arm effort by up to two-thirds The exoskeleton is worn like a backpack, with an electric motor mounted on the back. On both sides, thin, high-strength wires run from the back over the shoulders to the front, where they are attached to elbow-pad-like supports. When the motor pulls on the wires, they take over part of the work normally performed by the biceps muscle in the upper arm. A single-arm robot assists the person wearing the exoskeleton. When the robot picks up and hands over an object, it weighs it and wirelessly transmits its weight to the exoskeleton. The cables, which in their default setting precisely compensate for the weight of the wearer’s arms, then provide additional assistance proportional to the weight of the object. Recent studies show that this configuration can reduce muscular effort by up to 65%. “To handle asymmetrical components as well, the robot also determines their center of mass,” explains researcher Masiero. “The advantage is that one arm can receive more support than the other, making it possible to balance uneven loads.” The shoulder exoskeleton also works well on its own, but a bit differently: The most common research method is to measure muscle activity in the upper arm and use this to determine how much additional assistance is required at that moment. This method is fairly accurate, with an error margin of only 0.5 to 1 kilogram. However, the sensors must be attached to the upper arm before each use of the exoskeleton, which is rather impractical in a factory. Cobot: Programming by Demonstration The single-arm robot is a typical seven-joint cobot. This makes it mobile and flexible, while its ability to automatically slow down as it approaches people makes it safe enough for use in a research setting. The combination of the exoskeleton and the robotic arm offers particular advantages for industry: “We were able to demonstrate not only that we can specifically ease the physical workload of factory workers, but also how easy it is to teach the cobot new tasks,” explains Prof. Masia. “We can program it by guiding the robotic arm. Not a single line of code is needed. That’s a huge advantage over many robots currently used in factories, which operate behind safety barriers and away from human workers.”

TUM at the Zamanand Festival

At the AHA Knowledge Islands, exhibits, hands-on activities, demonstrations, and talks brought science to life for a broad audience. Everyone was invited to try things out, do crafts, and interact with the researchers. TUM’s topics ranged from health and sustainability to robotics and space exploration. Short presentations were also part of the program and facilitated direct dialogue with visitors.

ERC Advanced Grants awarded to Prof. Suyu and Prof. Saur

ERC grants are awarded in a number of categories. Advanced Grants are intended for established researchers who have achieved significant scientific accomplishments over the past ten years. The funding amount for these grants is up to €2.5 million each.

Using mini-organs to better understand aging

Awards presented at TUM Diversity Day 2026

TUM Lead with Care Award: Honoring exemplary leadership With the TUM Lead with Care Award 2026, TUM recognizes leaders who actively contribute to reconciling professional and family life – a crucial factor in attracting and retaining excellent talent. The awardees set an example for an appreciative workplace culture. This year’s awardees are: Dr. Astrid Bauer (Head of the Campus Office Weihenstephan, TUM School of Life Sciences)   Prof. Dr. Lars Mägdefessel (Professorship of Molecular Vascular Medicine, TUM School of Medicine and Health)   Prof. Dr. Doris Holzberger (Professorship of Psychology of Teaching and Learning, TUM School of Social Sciences and Technology) Congratulations on this award! The prize is awarded by the TUM Family Service and is endowed with 1,000 euros each.  Further information on the TUM Lead with Care Award  

TUM expands its venturing activities to Berlin

The aim of the strategic partnership is to establish a leading European network for startups in the field of defense and security. Linking Germany’s two largest startup communities is intended to provide founders with access to the resources of both locations – Munich and Berlin. Together with its TUM Venture Labs, TUM brings direct access at Marienpark to relevant stakeholders – from academia, politics and federal authorities to venture capital, security and defense companies, operators of critical infrastructure and civil protection organizations. This will help bring deep tech innovations into practice more quickly.

Lab-on-a-Chip Platform Shows How Immune Cells Attack Cancer Cells

Established laboratory tests mainly capture average values across many cells and show, for example, how many cancer cells survive after contact with immune cells. What happens in detail—how each cell reacts and interacts with others—remains hidden. However, to better understand the effectiveness of immunotherapies, the precise timing of a cell-cell interaction is often crucial: when contact, activation, and ultimately, the killing of the cancer cell occur.