We are happy to announce that Dr. Jelle Demanet from Spatial Dynamics, Prof. Song Gao from University of Wisconsin-Madison, Prof. Georg Gartner from TU Wien, Dr. Rohit Singh from Esri R&D Center India and Dr. Tony Vanderstraete from Digitaal Vlaanderen will be our keynote speakers at GeoAI 2026.
Jelle Demanet – Spatial Dynamics: Linking Human Behavior to Spatial Systems through Emerging Interfaces
The rapid evolution of spatial computing is fundamentally transforming how humans interact with digital systems. Emerging interfaces—ranging from extended reality (XR) and smart glasses to sensor-rich environments—are no longer passive visualization tools, but active layers through which humans perceive, interpret, and act within spatial data ecosystems. In parallel, Spatial AI and GeoAI are increasingly capable of modeling and understanding the physical world with high precision. However, a critical dimension remains underrepresented: the human behavioral component within these spatial systems.
This talk argues that the next frontier of GeoAI lies not only in modeling environments, but in integrating human behavior as a core data layer. As interfaces become more immersive and intuitive, they offer a unique opportunity to capture rich, multimodal behavioral signals—such as movement patterns, gaze behavior, decision-making processes, and interaction strategies—in real time and within ecologically valid contexts.
Building on this premise, we introduce the Spatial Dynamics platform, a behavioral intelligence infrastructure designed to systematically link human behavior to spatial digital applications. By embedding users in controlled yet realistic virtual and mixed reality environments, the platform enables the observation and quantification of human competencies and cognitive processes in direct interaction with spatial tasks.
The integration of behavioral data into spatial AI systems opens up transformative applications across multiple domains. In psychology, it enables more ecologically valid assessments of cognition and behavior. In product design and architecture, it allows for data-driven optimization of user experience and spatial interaction. In human resources, it facilitates objective, behavior-based skill assessment and job matching. In rehabilitation and education, it supports adaptive, personalized interventions grounded in real user behavior.
Ultimately, this contribution positions behavioral intelligence as a necessary extension of GeoAI, advocating for a shift towards truly human-centric spatial systems. By closing the loop between environment, interface, and human behavior, we can move towards more adaptive, inclusive, and effective spatial technologies.
Jelle Demanet is a cognitive neuroscientist, entrepreneur, and innovation leader working at the intersection of artificial intelligence, spatial computing, and human behavior. He is the CEO of Spatial Dynamics, a spin-off from imec, Ghent University, and Howest. The company develops a behavioral intelligence platform that captures and analyzes real human behavior in immersive environments, enabling objective insights into skills, cognition, and decision-making in spatial contexts.
With a background in cognitive neuroscience and experimental psychology, his work focuses on understanding how humans perceive and act within complex environments, using multimodal data such as movement, gaze, and interaction patterns to link human behavior to spatial AI systems.
Jelle is Chief Innovation Officer at ENT.A, where he contributes to innovation strategies across media and entertainment, as well as emerging technologies. He is co-founder and board member of XR Valley, and serves as Chairman of the Board of Image3D Europe, the organization behind UnitedXR, one of the leading XR conferences in Belgium.
Across his research and entrepreneurial activities, he explores how emerging interfaces—such as extended reality, digital twins, and sensor-driven environments—can enable more human-centric systems. His work spans applications in industry, human resources, education, healthcare, architecture, product design, and media & entertainment, with the goal of creating adaptive, inclusive, and data-driven environments that better align with human capabilities.
Song Gao – GeoAI: From Geospatial Questions to Intelligence
Geospatial artificial intelligence (GeoAI), a fast-growing interdisciplinary field, integrates geographic data and knowledge with AI technologies to address scientific challenges in human-environmental systems and earth sciences. It focuses on enhancing machines’ spatial intelligence to improve perception, reasoning, GIS workflow automation, knowledge discovery and mapping of geographic phenomena. This talk will introduce the core design of spatially-explicit GeoAI models, benchmarking and fine-tuning large language models (LLMs) for artificial GIS analyst and wildfire spread simulation, as well as addressing some of challenges and issues associated with GeoAI for social good.
Dr. Song Gao is an H.I. Romnes Professor with tenure in Geographic Information Science at the University of Wisconsin-Madison, where he leads the Geospatial Data Science Lab. He was elected as a Fellow of the American Association of Geographers (AAG) in 2024. He received his PhD degree from the University of California, Santa Barbara. His main research interests include Geospatial Data Science, GeoAI and Human Mobility. He is the lead editor of the GeoAI Handbook and the author of over 100 peer-reviewed research articles, published in prominent journals such as Science, PNAS, Nature Communications, and IJGIS. He is the PI of multiple research grants from the National Science Foundation and industry partners. He currently serves as the Associate Editor of International Journal of Geographical Information Science (IJGIS). Dr. Gao was the recipient of the UCGIS Early/Mid-Career Research Award and among the Web of Science Top 1% global highly cited researchers list.
Georg Gartner – From Reading Maps to Being Read by Maps
In an era when artificial intelligence is not only augmenting but reimagining the geospatial sciences, cartography stands at the threshold of a profound renaissance. From reading maps to being read by maps serves as a guiding metaphor, capturing a fundamental shift: maps are no longer passive artifacts we interpret, but intelligent systems that interpret us through streams of spatial data, real-time sensors, and adaptive algorithms. Today’s GeoAI landscape is defined by converging trends—autonomous agents capable of generating and executing spatial workflows, foundation-scale Earth models that fuse multi-source geospatial data into unified representations, and automated mapping pipelines that process vast datasets in minutes rather than weeks.
This presentation invites audiences to step back from individual tools to consider the broader intellectual and societal terrain of this transformation. GeoAI is rapidly becoming default infrastructure rather than experimental add-on, supported by pre-trained models, edge processing, and explainable outputs that help users understand why a prediction was made. Simultaneously, real-time geospatial applications, expanding sensor ecosystems, and increasingly accessible AI-powered geo-platforms are democratizing spatial intelligence across sectors and scales.
Yet as maps begin to “read” us, issues of ethics, transparency, bias, and accountability move from the margins to the core of cartographic practice. Trustworthy GeoAI requires not only technical excellence but principled design—systems that are interpretable, equitable, and socially responsible. By weaving together technological acceleration with critical reflection this presentation frames GeoAI as both a scientific evolution and a civic responsibility, outlining a future in which AI-enabled cartography deepens understanding, supports just decisions, and strengthens public trust in geographic knowledge systems.
Georg Gartner is an Austrian cartographer and geographer. He studied geography and cartography at the University of Vienna and later received his PhD and habilitation from the Vienna University of Technology. He is Professor of Cartography and Geo-Mediatechniques at TU Wien, where he leads the Research Unit Cartography. He has been heavily involved in international cartographic institutions. He served as President of the International Cartographic Association (ICA) from 2011 to 2015 and was re-elected President for the term 2023–2027. He also plays a central role in the International Master of Science in Cartography—a joint program of TU Wien, TU Munich, TU Dresden, and the University of Twente. In addition, he has contributed significantly to scientific publishing as Editor of theLecture Notes on Geoinformation and Cartography as well as the Journal of Location-based Services. In recognition of his contributions, he has received numerous awards and honours. These include Honorary Doctorates from Eötvös Loránd University (Budapest), the ICA Honorary Fellowship, and an Honorary Professorship from the Chinese Academy of Surveying and Mapping.
Rohit Singh – GeoAI in the Age of Foundation Models
Geospatial foundation models are redefining how we understand and analyze the Earth, unlocking new capabilities from vast, heterogeneous, and largely unlabeled geospatial data. In this keynote, Rohit Singh examines how GeoAI is evolving to incorporate these different model types into analytical workflows.
The talk brings together three emerging model classes: remote sensing foundation models, location embedding models, and geospatial vision-language models. It covers approaches ranging from self-supervised learning on multispectral and SAR imagery, to high-dimensional representations of place, to systems that combine imagery with natural language for spatial reasoning. Together, these approaches broaden the scope of what can be done in Earth observation and geospatial intelligence.
Drawing on a survey of leading models and research, the keynote outlines current capabilities, ongoing challenges, and practical ways to integrate these models into real-world geospatial workflows. It also offers a look at Esri’s work in this area.
By treating embeddings as a core form of geospatial data, this talk points to a shift toward embedding-driven analysis, prediction, and reasoning, with potential applications in areas such as environmental monitoring, urban systems, and beyond.
Rohit Singh is Director of Esri’s R&D Center in New Delhi, where he leads the design and development of Geospatial AI capabilities across the ArcGIS platform. His work focuses on advancing GeoAI, foundation models, and scalable spatial analytics to power next-generation geospatial applications. He has played a key role in building core ArcGIS technologies, including the ArcGIS API for Python, the ArcGIS Java Engine API, and the platform’s Linux enablement. Rohit holds an MS in Computer Science with a specialization in AI from Georgia Tech and is an alumnus of IIT Kharagpur.
Tony Vanderstraete – Forward-looking policy making, how can GeoAI help answer societal challenges?
Flanders increasingly faces complex societal challenges in a rapidly evolving world. A forward-looking digital policy not only ensures that data and digital technologies can be leveraged to address these societal challenges, but also provides the necessary agility to respond to an ever changing context. How can GeoAI be part of the solution?
Tony Vanderstraete graduated in 1999 from Ghent University with a Master’s degree in Geography and obtained a PhD in Sciences from the same university in 2007. Following an academic career focused on the use of remote sensing to address societal challenges, Tony joined the Flemish government. There, until 2020, he was responsible for innovative data acquisition, data dissemination and business development. Since then, Tony serves as Head of Digital Strategy and leads, among other initiatives, the AI Expertise Centre at Digitaal Vlaanderen which ensures the responsible use of AI within the Flemish government.