About the project
Objective
- Careful assessment of what new information due to personal data of an individual has been memorised;
- Guidelines outlining what residuals of personal information in a ML model should be legally acceptable;
- Explore how the RTBF can be effectively implemented, ultimately in the context of ML systems.
Background
In the context of machine learning (ML), there is an urgent need to clarify what “deletion” under the Right to be Forgotten (RTBF) truly entails. As ML models generalize and internalize patterns from data, achieving complete removal of an individual data point remains a major technical challenge. We argue, however, that such full deletion often exceeds what the law actually requires. REFLECT-ML aims to explore how the RTBF can be effectively implemented, ultimately in the context of ML systems. We will address the disconnect between the abstract legal language of regulators and operationally founded technical theoretical measures that can be used to quantify the new information associated with individual data as well as the practical complexities in estimating those. Further, we will develop different technical approaches to control its memorization as well as exploring machine unlearning attempts.
Cross-disciplinary collaboration
The team consists of PI Oechtering (researcher in information theory, relevant for information quantification), PI Colonna (researcher in law, relevant knowledge related to the GDPR and AI Act), and PI Johansson (researcher in optimization, relevant for the design of efficient learning algorithms). The outlined work is a cross-disciplinary effort since information measures need to be legal compliant, legal requests need to be algorithmically feasible, and algorithms need to aim for the right objective.
About the project
Objective
Code generation with AI is making very fast progress. Nonetheless, it suffers from a fundamental drawback: it is unreliable, and users developers cannot trust the generated code, which is a major blocker. Our project TRUST-SW aims to solve this problem: AI will generate code alongside verification materials using state-of-the-art formal verification. By doing so, project TRUST-SW will enable users to produce dependable, correct code with generative AI. The specific objectives are:
- Generate specification from human language using LLMs: Develop capabilities for LLMs to translate human language into specification with no or minimal human supervision.
- Empower LLMs with deductive, explainable reasoning: Equip LLMs with deductive reasoning skills to debug and validate code based on natural language instructions and novel techniques.
- Embed and evaluate real-world constraints into code generation: Ensure that generated code aligns with practical requirements, handling complex constraints of real-world applications.
Background
Recent advances in generative AI, particularly large language models, have made it possible to automatically generate software from natural language prompts. While these tools significantly accelerate software development, the generated code may contain subtle bugs, security vulnerabilities, or logical errors that are difficult to detect. At the same time, formal verification techniques can mathematically guarantee software correctness but often require significant expertise and manual effort. TRUST-SW addresses this gap by exploring how AI systems can work together with formal reasoning and verification tools to automatically generate software that is not only functional but also provably correct and trustworthy.
Cross-disciplinary collaboration
TRUST-SW brings together expertise from several research areas, including artificial intelligence, formal verification, programming languages, and software engineering. The project combines advances in large language models with rigorous methods from formal methods and automated reasoning. This interdisciplinary collaboration enables the development of novel approaches where AI-generated code can be systematically verified and validated, bridging the gap between machine learning–based code generation and mathematically grounded software assurance.
PIs: Marco Chiesa, Martin Monperrus, Mariano Scazzariello.
About the project
Objective
Study the issue of outliers’ privacy from an information-theoretic point of view, propose an adapted privacy notion, such as pointwise maximal leakage, to solve it and design sanitising mechanisms in the light of these principled insights.
Background
The most popular privacy measure, differential privacy, can only protect outliers at the cost of destroying accuracy. Its relaxation, metric differential privacy, fails to guarantee the privacy of such isolated points.
About the Digital Futures Postdoc Fellow
Arnaud Grivet Sébert completed his PhD in CEA List, Gif-sur-Yvette, France, under the direction of Renaud Sirdey and the co-supervision of Cédric Gouy-Pailler. He proposed approaches that combine differential privacy and homomorphic encryption to protect the training data privacy in distributed machine learning.
He then worked on the privacy of textual data, and especially outliers, as a post-doctoral researcher in LIX (Laboratoire d’Informatique de l’Ecole Polytechnique), Palaiseau, France, with Catuscia Palamidessi and Sonia Vanier, and in Macquarie University, Sydney, Australia, with Annabelle McIver and Mark Dras.
He is now starting a post-doctoral contract in KTH, funded by Digital Futures and supervised by Tobias Oechtering and Martina Scolamiero. He is especially interested in the theoretical aspects of privacy, but also in its relations with other ethical properties like frugality, robustness, fairness.
Main supervisor
Tobias Oechtering, KTH
Co-supervisor
Martina Scolamiero, KTH
About the project
Objective
The project primarily addresses trustworthy AI deployment for mission-critical robotic systems operating in industrial and adversarial environments.
EdgeWise aims to redefine how Vision-Language-Action (VLA) models are deployed, executed, and secured for next-generation humanoid robots operating in connectivity-constrained and adversarial environments.
The main objectives are to:
- Design a novel three-tier VLA deployment architecture (on-device, edge, cloud) to reduce end-to-end control loop latency and increase resilience in unstable network conditions.
- Enable efficient multi-tenant edge model serving by compressing fine-tuned models into lightweight adapters that share a common base model, significantly improving resource utilization.
- Introduce verifiable and interpretable execution planning, where robots generate structured code from natural language instructions and apply formal verification methods to ensure correctness and safety before execution.
Background
Humanoid robots powered by Vision-Language-Action models have the potential to transform healthcare, construction, search-and-rescue, and other labor-intensive or hazardous domains. However, current VLA architectures rely heavily on cloud-based reasoning models, which introduce latency and depend on stable connectivity. This makes them unsuitable for many real-world scenarios such as disaster zones, underground mining, or remote medical environments.
Moreover, VLA systems often generate opaque action tokens that are difficult to interpret or verify, raising serious safety and security concerns. Recent studies show that language-model-controlled robots can be manipulated into performing unsafe actions, highlighting the urgent need for transparency and formal guarantees.
EdgeWise addresses these limitations by combining edge computing, efficient model sharing mechanisms, and formal verification techniques to enable secure, resilient, and low-latency robotic AI systems.
Crossdisciplinary collaboration
EdgeWise is a collaboration between RISE and KTH, combining expertise in:
- Distributed systems and edge/cloud orchestration
- Machine learning systems and large-model deployment
- Formal verification and secure software systems
- Robotics and real-world experimental testbeds
The project integrates systems research, AI model optimization, networking (5G/6G), and formal methods. The experimental platform includes robotic systems connected to a private 5G infrastructure and research data center resources, enabling controlled evaluation of latency, resilience, and safety mechanisms.
PI: Joakim Eriksson, RISE
Co-PI: Marco Chiesa, KTH
About the project
Objective
The objective of this project is to develop an AI-enabled, fully self-powered, and biodegradable wound-healing patch that accelerates tissue regeneration and enables continuous monitoring of post-cardiac-surgery wounds. The system combines triboelectric nanogenerators (TENGs) for bioelectric stimulation with AI-based image analysis to provide personalized, sustainable, and real-time wound care without external power sources.
Background
Post-cardiac-surgery wounds face high risks of infection, delayed healing, and limited continuous monitoring. Existing wound-care technologies rely on external power sources, are costly, and lack portability. Triboelectric nanogenerators (TENGs) offer a promising alternative by harvesting biomechanical energy from natural body movements to deliver gentle bioelectric stimulation.
This project integrates biodegradable hydrogels with antibacterial and anti-inflammatory properties and AI-driven wound image analysis to assess healing stages such as inflammation, proliferation, and remodeling. The approach reduces electronic waste, enables continuous monitoring, and supports faster, safer recovery through sustainable digital healthcare solutions.
About the Digital Futures Postdoc Fellow
Swati Panda is a postdoctoral researcher at the Department of Biomedical Engineering and Health Systems at KTH, specializing in biocompatible and biodegradable energy-harvesting devices for self-powered healthcare applications. She holds a PhD in Robotics and Mechatronics Engineering from DGIST, South Korea. Her research focuses on triboelectric and piezoelectric nanogenerators, biodegradable/biocompatible polymers, and AI-based signal and image processing for healthcare sensing.
She has extensive experience in material synthesis, device fabrication, in-vivo experimentation, and self-powered health monitoring systems. Through her work, she aims to develop sustainable, wearable, and smart healthcare technologies that improve patient outcomes while reducing environmental impact.
Main supervisor
Seraina Dual, Assistant Professor, Department of Biomedical Engineering and Health Systems, KTH Royal Institute of Technology.
Co-supervisor
Erica Zeglio, Assistant Professor, Department of Chemistry, Stockholm University.
About the project
Objective
WWF Sweden’s One Planet Plate (OPP) offers chefs, caterers and consumers a practical guide to climate- and biodiversity-aligned meals, underpinned by a “cradle-to-plate” database and a 0.5 kg CO₂e/meal threshold. Despite early uptake by more than twenty restaurants and public institutions, OPP faces challenges, such as data and methodological inconsistencies, trust issues and scale-up barriers.
Addressing these challenges, our project aims to shed light on the following research questions:
- How and under which conditions can the OPP software diffuse and drive transitions toward economically and environmentally sustainable business models of restaurants?
- How can the methodological foundations and user-facing functionalities of the OPP tool be critically evaluated and improved, e.g., through uncertainty communication or AI integration, to better support climate- and biodiversity-aligned food practices?
- How and under which conditions can stakeholder engagement and cooperation foster mutual learning and inform the future direction of data-driven sustainability tools in food services?
Background
As datafication and artificial intelligence reshape decision-making, including in food services,One Planet (OPP) initiative offers an early real-world testbed. In our project, we will providean in-depth study of OPP – one of frontrunners among data-driven sustainable food initiatives worldwide – advancing understanding of the potential of digital tools to foster climate- and biodiversity-aligned consumption.
We apply a transdisciplinary lens – exploring methodological robustness, user experience, scalability and business feasibility – through close collaboration between researchers, tool developers, data providers, chefs and other users.
Crossdisciplinary collaboration
The researchers in the team represent the Department of Industrial Economic and Management (INDEK) at ITM/KTH and the Department of Sustainable Development, Environmental Science and Engineering (SEED) at ABE/KTH as well as KTH Food Centre.
About the project
Objective
This project aims to develop a DT-based risk-informed decision support methodology, specifically addressing the risk issue associated with stability and manoeuvrability loss of a scaled wind-assisted propulsion ship system. By integrating the DTs into Dynamic Risk Analysis (DRA) methodology, real-time risk estimation can be achieved, enabling proactive risk identification and control. Building on these real-time risk insights, a risk-informed decision support model will be trained to optimise sail attack angle adjustments, ensuring the scaled ship operates within a low-risk state.
Background
Digitalisation and decarbonisation are identified as the key transformative forces shaping the future of shipping, driving the adoption of marine green technologies. Therein, wind-assisted propulsion ship is recognised as one of the most promising solutions for green shipping. However, the deployment of large-scale sails in the wind-assisted propulsion system (WAPS) significantly alters ship’s weight and load distribution, increasing the inherent risk of stability and manoeuvrability loss, which threatens crew lives, assets and the marine environment.
Additionally, the continuous adjustments of sail direction to adapt to changing wind conditions during navigation introduce further uncertainties, amplifying the fluctuations in risk and highlighting the challenges in managing the dynamic risk levels associated with these systems. Therefore, achieving resilient and smart green shipping, which involves risk-interactive and adaptive automated operations, requires real-time, risk-informed decision support integrated with high-fidelity digitalisation. In this context, Digital Twins (DTs), by integrating the virtual and physical worlds, enable the real-time monitoring of operational scenarios and timely data analysis to head off issues before they arise, which holds great potential to enhance the risk-informed decision support yet still remains largely untapped.
About the Digital Futures Postdoc Fellow
Yue Han holds a Ph.D. in Design and Manufacture of Ship and Ocean Structure from Dalian University of Technology in China, where she specialised in the development of risk assessment methodologies and frameworks. With 7 years of research experience in this field, she has been working on dynamic risk analysis and intelligent decision-making of risk control strategies for marine structures such as ships, offshore installations, and their equipment. Now Yue is working on addressing the new hazards and failure mechanisms emerging from the recent implementation of green technologies and digital technologies in the maritime industry.
Main supervisor
Abbas Dashtimanesh
Co-supervisor
Jelena Zdravkovic and Giuseppe Belgioioso