A catalyst for innovation
The JENAINNOVATION transfer centre
Seeking solutions
Latest research from the Light, Life and Liberty research profiles
Physicist Dr Elena Goi in her research laboratory. She heads a junior research group aiming to develop applications for neuromorphic optical systems.
How can we gain knowledge without wasting resources and without drawing the wrong conclusions? Researchers in Jena are developing approaches that consider technology and society together. Three examples.
Text: Sebastian Hollstein and Ute Schönfelder
Artificial intelligence is increasingly shapingour everyday lives
AI answers customer enquiries via chatbots, generates text, designs images and carries out a wide range of tasks autonomously. However, the computers required for these tasks are becoming larger, more expen-
sive and more energy-intensive. Modern AI models are increasingly pushing current hardware to its limits—a problem that will only intensify in the coming years. Physicists in Jena are developing optical computers in the new »PicPhotMat—Structured Materials for Picophotonic Analog Computing« project funded by the Federal Ministry of Research, Technology and Space. These computers operate with light rather than electricity, using computing units as small as the atoms that make up a crystal. This concept could significantly increase AI computing performance and open the door to an extremely energy-efficient, compact and sustainable alternative to contemporary supercomputers.
We want to develop AI hardware that consumes less energy while delivering significantly higher
performance than existing systems.Elena Goi
»Modern AI methods are based on neural networks and achieve brain-like performance in tasks that are diffi-
cult for traditional computers but easy for humans,« explains Dr Elena Goi, the project leader and head of the new junior research group associated with the project, the »NanoPico Photonic Computing Group«, at the Institute of Applied Physics. »Optical systems are ideally suited for rapidly processing the large amounts
of data that neuromorphic—i.e. brain-inspired—systems handle. They can perform complex mathematical operations on large matrices—which are the fundamental building blocks of AI algorithms—with far greater energy efficiency than conventional processors, thereby outperforming conventional silicon-based electronic technologies.«
Miniaturizing optical components while implementing flexible designs
Despite their great potential, these neuromorphic optical systems are not yet ready for widespread use. One of the main reasons is that, to date, it has been difficult to miniaturize them and integrate them in large numbers on a computer chip. At present, artificial optical neurons still require a great deal of space on a chip, even when state-of-the-art manufacturing techniques are deployed. While so-called metamaterials do ena-
ble very compact elements for information processing, these are typically static and must be tailored for each application. For versatile AI systems that are intended to respond flexibly to different tasks, this is too resource-intensive. This is precisely where »PicPhotMat« comes in, seeking to miniaturize optical components and make them flexible enough to be used as widely as modern electronic chips.
Lower energy consumtion—enhanced performance
»Our project applies the latest findings from the field of so-called picophotonics, a branch of research that studies interactions between light and matter at scales a million times smaller than the diameter of a human hair,« explains the physicist. »The interaction of light and matter on this scale can strongly influence how
light behaves and performs. This opens up new possibilities for the development of highly compact, energy-efficient and scalable optical neuromorphic systems.«
With this specialized technology, the »PicPhotMat« project is paving the way for sustainable and high-performance AI hardware for the future. The researchers aim to develop chips that consume less energy while delivering significantly higher performance than current systems. Furthermore, the project is expected to provide new insights into how light and matter interact at the smallest scale—knowledge that could open up entirely new technical possibilities over the long term.
Medical research needs models
Professor Alexander Mosig is working at a laminar flow cabinet in a laboratory.
Image: Jürgen Scheere (University of Jena)Prof. Dr Alexander Mosig and his team are developing »organ-on-a-chip« models at Jena University Hospital: human cells are grown in small chambers on microscope slides to form functional tissue, on which a wide range of experiments can be carried out. For Mosig, these »artificial tissues« and »mini-organs« represent a genuine alternative to animal testing—and not solely for ethical reasons. First and foremost, this is because organ models often allow crucial questions to be answered more precisely.
In the Cluster of Excellence »Balance of the Microverse«, of which he has been a member since the start of 2026, Mosig is focusing on the human microbiome and investigating how pathogens colonize the human gut. »How do the billions of
bacteria, viruses and fungi interact with our bodies? When do these interactions promote inflammation or infection, and when do they make us more resilient, for example against pathogens or in metabolic disorders?« asks Mosig. The chal-
lenge, he says, is not only to understand the underlying mechanisms between microorganisms and human cells, but also to harness them in such a way that the prevention or treatment of diseases becomes possible.
No model is without limitations
Observational studies involving patients clearly have their limitations. »We cannot simply infect healthy people with pathogens and see what happens,« says the biochemist. Consequently, medical research needs good models. Animal models do have their place, but they all inevitably omit certain variables. »In animal models, we generally use animals of a specific age, with a particular diet or genetic background—and this always represents only a highly constrained condition,« he says. What works in animal models is therefore not automatically transferable to humans. Mosig therefore relies on a portfolio of several complementary model systems. More important than the question »Which model is the best?« is understanding how findings from different models complement one another.
The »organ-on-a-chip« model, too, captures only certain aspects. »But compared to animal models, we have the advantage of being able to examine human tissue and human immune cells under controlled conditions. And not just as a static snapshot, but also as a dynamic system,« says Alexander Mosig.
We don’t just want to replace animal models. We want to create better models!
Alexander Mosig
The chips, produced by a start-up company spun off from Mosig’s team, consist of various compartments connected by a permeable membrane, through which fluid flows via tiny channels. This creates, for example, a model of the intestinal mucosa that can be colonized by microorganisms and interacts with circulating immune cells via a simulated blood vessel system. Because the system is placed on a microscope-compatible slide, researchers can observe in real time how tissue, pathogens and immune cells behave.
Models for specic human diseases
Organ-on-a-chip models offer a further decisive advantage whenever there are no suitable animal models available. Mosig’s team is therefore working, amongst other things, within the Cluster of Excellence »Balance of the Microverse« on projects focusing on infectious diseases that affect only humans.
One such disease is cholera, a diarrhoeal illness. Its causative agent, Vibrio cholerae—a bacterium that infects only humans—spreads primarily in areas where clean drinking water is lacking. Together with Prof. Kai Papenfort, Mosig’s group is developing an intestine-on-a-chip model in which the cholera pathogen can be cultured. This allows the team to observe it, deliberately alter the growth conditions and test new therapeutic approaches.
In a second project, Alexander Mosig is focusing on a common seasonal virus—the norovirus, which is one of the most common causes of gastrointestinal illness in humans, with symptoms such as severe vomiting and diarrhoea. »For most people, this is certainly very unpleasant, but they also recover quite quickly,« says Mosig. »But for young children, the elderly or people with underlying health conditions, the infection can be life-threatening.«
There is currently no effective treatment for the infection. With this pathogen, too, researchers face the dilemma that the human norovirus replicates almost exclusively in humans—there are no suitable animal models. Mosig’s team therefore uses a three-dimensional mini-tissue on the chip that mimics intestinal functions. This allows the virus to be cultured and the effects of antibodies or drug candidates to be systematically tested.
New standards in drug development
Alexander Mosig’s conclusion: Animal testing is unlikely to disappear altogether any time soon. »However, where »organ-on-a-chip« models deliver results more quickly and, above all, more accurately, they should become the new standard.« In the US, it has been possible for several years now to avoid animal testing in drug development in order to promote the use of animal-free methods. In the EU, animal testing is currently still mandatory under the European Medicines Agency (EMA) before new active substances can be tested in clinical trials on humans.
What will sustain us tomorrow—Imaginamics
In order to shape the future, a society must agree on what that future should actually look like. Fundamental to the creation of shared ways of imagining the world—alongside all processes of negotiation—is social imagination. Through this ubiquitous yet largely invisible phenomenon, individual images and narratives within groups of people come together to form shared imaginary spaces. It creates a basis for understanding, thereby holding societies together, but it can also give rise to potential for exclusion and division.
Scientists are investigating exactly how social imagining works in the new Cluster of Excellence »Imaginamics: Practices and Dynamics of Social Imagining«. »We want to understand how shared imaginations emerge, circulate and generate new dynamics,« explains Prof. Dr Christina Brandt. The historian of science is one of three project spokespersons, alongside art historian Prof. Dr Johannes Grave and historian Prof. Dr Joachim von Puttkamer.
We can gain a better understanding of many of the crises we are currently facing by looking at social imagination.
Christina Brandt
The first step is to examine the practices of diverse forms of social imagining. In addition to traditional visual representations such as paintings or photographs, people also make use of linguistic devices such as metaphors and narratives. In research, visual thought experiments such as Schrödinger’s cat are also used to illustrate a scientific problem. On social media, memes are used to express how users feel. Images are everywhere—but that alone does not mean that what they express is immediately understood and shared.
»Imagination is an individual faculty—a group does not possess any particular power of imagination; in this respect, the existence of social imaginations requires explanation,« says Johannes Grave. »In a social context, the meaning of images can be enriched and shifted, giving rise to ideas that were not specifically intended.« This dynamic probably stems less from the image itself than from how people engage with it and the accompanying circumstances. Michelangelo’s famous statue of »David«, for example, originally embodied the self-assertion and defensive strength of the Republic of Florence. Anyone viewing the statue today no longer sees it as a political symbol, but as an icon of Renaissance art. This is due to changes in the political context, the canonization of the period, and the statue’s relocation: In the 19th century, the sculpture was moved from the public square in front of the town hall to the Art Academy’s museum.
Figurative language is also subject to change, as an example from the world of science shows: geneticists coined the term »genome editing« to summarize methods for the precise modification of DNA in specialist publications. Media coverage picked up on the phrase and supplemented it with further imagery such as »genetic scissors«, which tends to lead the public to conceive of complex laboratory processes in simplified terms, thereby failing to realistically assess the possibilities and risks of genetic research.
We are increasingly confronted with the question of whether we all still share the same reality.
Johannes Grave
Observing, understanding and educating the public about such processes is of immense importance to society. »We can better understand many of the crises currently facing us, such as the questioning of democracy by various groups, by looking at social imagination,« says Christina Brandt. »For the imaginary exerts a more subtle form of power, which only becomes apparent once one understands the mechanisms that generate it.« And this collective lens of imagination is starting to falter. »We are increasingly confronted with the question of whether we all still share the same reality,« says Johannes Grave. »It is possible that manipulators are not only spreading misinformation, but are interfering directly with the inner workings of imagination.«
It is no coincidence that a Cluster of Excellence focusing on the foundations of social imaginaries is being established in Jena of all places. For some time now, representatives from the humanities and social sciences have been working closely together here. »Imaginamics« offers an opportunity to highlight the important role of the cultural and humanities disciplines in tackling key questions in the social sciences.
»In literary studies, art history and intercultural communication, researchers are constantly engaged with storytelling, and with the creation, perception and use of images. Social imagination has long been a focus here,« explains Johannes Grave. »Within the framework of the cluster, we aim to place this practice in a broader social context.« For it can contribute not only to our understanding of Fontane’s »Effi Briest«, Caspar David Friedrich’s »Monk by the Sea« or Francis Ford Coppola’s »The Godfather«, but also to the further development of social theories—and to a common foundation upon which we can envision our future.
