Periodic Reporting for period 1 - HURO (Leading in the brave new world: human - robot dynamics)
Période du rapport: 2024-10-01 au 2025-09-30
The interface between physical and digital domains is rapidly fading, raising questions about the ramifications that technological innovations have for organisations. Despite offering plenty of opportunities, a more digitalised world of work changes the nature and flow of information, inevitably creating novel challenges for individuals, organisations and society. Therefore, this project aims to tackle the following questions: 1) How can we use AI to measure individual characteristics and lying at-a-distance accurately? 2) Can at-a-distance assessment methods based on AI reduce bias in the assessment process? 3) What behavioural strategies can we develop to ensure a high organisational acceptance of AI/Robotics measures? The proposed project will provide important theoretical, methodological, and practical implications by harnessing robust empirical research.
During the first phase, the project implemented the core scientific activities across four interrelated studies addressing leadership, human–AI, and human–robot collaboration.
Study 1 was fully completed and focused on identifying leadership capabilities required for effective human–AI and human–robot teamwork. Thirty semi-structured interviews with leaders working in hybrid teams were conducted and analysed using classical thematic analysis and computational topic modelling techniques. This study resulted in a validated set of leadership capabilities specific to AI-enabled work contexts and constitutes the conceptual foundation for the subsequent studies.
Study 2 examined human–robot collaboration through a controlled laboratory experiment. Participants collaborated with a humanoid robot under different behavioural conditions to assess performance, preferences, willingness to interact, and physiological stress indicators. Data collection was completed during the reporting period, generating a rich experimental dataset that is currently under analysis.
Study 3 focused on the development and validation of AI-based models for assessing leadership attributes at a distance. A large-scale field study involving approximately 2,000 participants was completed. Supervised deep learning models using natural language processing techniques were developed and tested to infer personality traits, charisma, and cognitive ability from speech data. The models were evaluated using out-of-sample testing, demonstrating the technical feasibility and scientific contribution of the approach.
Study 4 builds on the results of Study 3 and investigates how individuals accept and respond to feedback originating from human versus AI sources. During the reporting period, the experimental design was finalised and data collection was initiated.
Overall, the project has successfully completed or advanced all planned scientific activities for the interim period and has achieved its main technical objectives in line with the Description of the Action.
Study 1 was fully completed and focused on identifying leadership capabilities required for effective human–AI and human–robot teamwork. Thirty semi-structured interviews with leaders working in hybrid teams were conducted and analysed using classical thematic analysis and computational topic modelling techniques. This study resulted in a validated set of leadership capabilities specific to AI-enabled work contexts and constitutes the conceptual foundation for the subsequent studies.
Study 2 examined human–robot collaboration through a controlled laboratory experiment. Participants collaborated with a humanoid robot under different behavioural conditions to assess performance, preferences, willingness to interact, and physiological stress indicators. Data collection was completed during the reporting period, generating a rich experimental dataset that is currently under analysis.
Study 3 focused on the development and validation of AI-based models for assessing leadership attributes at a distance. A large-scale field study involving approximately 2,000 participants was completed. Supervised deep learning models using natural language processing techniques were developed and tested to infer personality traits, charisma, and cognitive ability from speech data. The models were evaluated using out-of-sample testing, demonstrating the technical feasibility and scientific contribution of the approach.
Study 4 builds on the results of Study 3 and investigates how individuals accept and respond to feedback originating from human versus AI sources. During the reporting period, the experimental design was finalised and data collection was initiated.
Overall, the project has successfully completed or advanced all planned scientific activities for the interim period and has achieved its main technical objectives in line with the Description of the Action.
The project advances the state of the art by moving beyond predominantly conceptual or technology-centric approaches to AI and robotics, and by providing empirical, interdisciplinary evidence on how human–AI and human–robot collaboration actually functions in organisational settings. It delivers novel insights by combining qualitative leadership research, controlled experimental studies, and large-scale computational modelling to examine leadership capabilities, interaction dynamics, and feedback processes in hybrid human–AI teams.
Beyond existing literature, the project contributes scientifically validated tools and evidence that link leadership theory with human-centred AI and robotics, an area that remains underdeveloped despite rapid technological deployment. In particular, the project demonstrates how leadership behaviours, robot roles, and feedback mechanisms shape trust, acceptance, performance, and well-being in AI-enabled work contexts.
Potential economic and societal impacts include improved efficiency and reduced coordination costs through better-designed human–AI team configurations, as well as more inclusive and sustainable adoption of AI technologies in organisations. While quantitative economic impact will materialise at later stages, the project establishes the necessary evidence base for such outcomes.
Further uptake and success of the results will require:
1) continued research and validation in real-world organisational settings,
2) integration of findings into leadership training, skills development, and organisational guidelines,
3) alignment with emerging regulatory and standardisation frameworks for responsible and human-centred AI,
4) dissemination to policymakers, practitioners, and standard-setting bodies concerned with the future of work and AI governance.
These steps will enable the project’s results to move from research evidence toward broader organisational and policy uptake by the end of the fellowship and beyond.
Beyond existing literature, the project contributes scientifically validated tools and evidence that link leadership theory with human-centred AI and robotics, an area that remains underdeveloped despite rapid technological deployment. In particular, the project demonstrates how leadership behaviours, robot roles, and feedback mechanisms shape trust, acceptance, performance, and well-being in AI-enabled work contexts.
Potential economic and societal impacts include improved efficiency and reduced coordination costs through better-designed human–AI team configurations, as well as more inclusive and sustainable adoption of AI technologies in organisations. While quantitative economic impact will materialise at later stages, the project establishes the necessary evidence base for such outcomes.
Further uptake and success of the results will require:
1) continued research and validation in real-world organisational settings,
2) integration of findings into leadership training, skills development, and organisational guidelines,
3) alignment with emerging regulatory and standardisation frameworks for responsible and human-centred AI,
4) dissemination to policymakers, practitioners, and standard-setting bodies concerned with the future of work and AI governance.
These steps will enable the project’s results to move from research evidence toward broader organisational and policy uptake by the end of the fellowship and beyond.