Theodore Jensen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Theodore Jensen has a total of 23 co-authors. Here are the top co-authors by number of shared publications:
- Emil Coman
Health Disparities Institute, University of Connecticut, Hartford, CT 06030-7030, USA
3 shared publications - Mohammad Maifi Hasan Khan
Department of Computer Science and Engineering, University of Connecticut, USA
3 shared publications - Ross Buck
University of Connecticut
3 shared publications - Yee-Yin Choong
National Institute of Standards and Technology
3 shared publications Yusuf Albayram
Central Connecticut State University
3 shared publicationsAbdullah Al Fahim
Department of Computer Science & Engineering, University of Connecticut, Storrs, CT, USA
2 shared publications
Latest Publications
- 13 November 2025
- Published by National Institute of Standards and Technology (NIST)
Assessing risks and impacts of AI (ARIA) :
This document describes the procedure used for a pilot of NIST’s Assessing Risks and Impacts of AI (ARIA) evaluation: ARIA 0.1. Five organizations participated, submitting a total of 7 AI applications to be evaluated. In this document, we first describe the design of the three evaluation scenarios (TV Spoilers, Meal Planner, Pathfinder) and the three testing levels (model testing, red teaming, field testing). We then discuss the methods used for assessment via dialogue annotation and tester questionnaires. Finally, we describe our approach to measuring validity of AI applications using measurement trees.
- 13 November 2025
- Published by National Institute of Standards and Technology (NIST)
Assessing risks and impacts of AI (ARIA) :
This document describes the procedure used for a pilot of NIST’s Assessing Risks and Impacts of AI (ARIA) evaluation: ARIA 0.1. Five organizations participated, submitting a total of 7 AI applications to be evaluated. In this document, we first describe the design of the three evaluation scenarios (TV Spoilers, Meal Planner, Pathfinder) and the three testing levels (model testing, red teaming, field testing). We then discuss the methods used for assessment via dialogue annotation and tester questionnaires. Finally, we describe our approach to measuring validity of AI applications using measurement trees.
- 30 May 2025
- Published by Springer Nature
- p. 86-105
Means to an End: Assessing the Validity of Trust Scales in Artificial Intelligence Vignettes
- 30 May 2025
- Published by Springer Nature
- p. 86-105
Means to an End: Assessing the Validity of Trust Scales in Artificial Intelligence Vignettes
- 30 May 2025
- Published by Springer Nature
(undefined)
- Open Access
- 16 October 2024
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society
- Vol. 7, 647-658
Reflection of Its Creators: Qualitative Analysis of General Public and Expert Perceptions of Artificial Intelligence
The increasing prevalence of artificial intelligence (AI) will likely lead to new interactions and impacts for the general public. An understanding of people’s perceptions of AI can be leveraged to design and deploy AI systems toward human needs and values. We conducted semi-structured interviews with 25 individuals in the general public and 20 AI experts in the United States (U.S.) to assess perceptions of AI across levels of expertise. Qualitative analysis revealed that ideas about humanness and ethics were central to perceptions of AI in both groups. Humanness, the set of traits considered to distinguish humans from other intelligent actors, was used to articulate beliefs about AI’s characteristics. Ethics arose in discussions of the role of technology in society and centered around views of AI as made and used by people. General public and expert participants expressed similar perceptions of AI, but articulated beliefs slightly differently. We discuss the implications of humanness-related beliefs and ethical concerns for AI development and deployment.
- 16 October 2024
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society
- Vol. 7, 647-658
Open AccessReflection of Its Creators: Qualitative Analysis of General Public and Expert Perceptions of Artificial Intelligence
The increasing prevalence of artificial intelligence (AI) will likely lead to new interactions and impacts for the general public. An understanding of people’s perceptions of AI can be leveraged to design and deploy AI systems toward human needs and values. We conducted semi-structured interviews with 25 individuals in the general public and 20 AI experts in the United States (U.S.) to assess perceptions of AI across levels of expertise. Qualitative analysis revealed that ideas about humanness and ethics were central to perceptions of AI in both groups. Humanness, the set of traits considered to distinguish humans from other intelligent actors, was used to articulate beliefs about AI’s characteristics. Ethics arose in discussions of the role of technology in society and centered around views of AI as made and used by people. General public and expert participants expressed similar perceptions of AI, but articulated beliefs slightly differently. We discuss the implications of humanness-related beliefs and ethical concerns for AI development and deployment.
- 26 March 2024
- Published by National Institute of Standards and Technology (NIST)
AI use taxonomy :
As artificial intelligence (AI) systems continue to be developed, humans will increasingly participate in human-AI interactions. Humans interact with AI systems to achieve particular goals. To ensure that AI systems contribute positively to human-AI interactions, it is important to examine human-AI tasks with an emphasis on human goals and outcomes. The AI Use Taxonomy aims to provide a flexible means of classifying how an AI system contributes to an outcome. The taxonomy sets forward 16 AI use “activities” which are independent of AI techniques and domains. Tasks are combinations of one or more AI use activities. Future research includes applying the taxonomy to better understand measurement challenges for each activity. The taxonomy can contribute to an improved understanding of the architecture of human-AI tasks and help to foster positive, human-centered interactions with AI systems and optimal outcomes in the following ways: • Provides common terminology for describing outcome-based human-AI activities independent of AI techniques and domains • Enables cross-domain insights based on shared human-AI activities • Highlights commonalities in measurement and evaluation needs across disparate AI techniques • Facilitates the development of use cases • Facilitates the evaluation of trustworthiness characteristics and usability.
- 26 March 2024
- Published by National Institute of Standards and Technology (NIST)
AI use taxonomy :
As artificial intelligence (AI) systems continue to be developed, humans will increasingly participate in human-AI interactions. Humans interact with AI systems to achieve particular goals. To ensure that AI systems contribute positively to human-AI interactions, it is important to examine human-AI tasks with an emphasis on human goals and outcomes. The AI Use Taxonomy aims to provide a flexible means of classifying how an AI system contributes to an outcome. The taxonomy sets forward 16 AI use “activities” which are independent of AI techniques and domains. Tasks are combinations of one or more AI use activities. Future research includes applying the taxonomy to better understand measurement challenges for each activity. The taxonomy can contribute to an improved understanding of the architecture of human-AI tasks and help to foster positive, human-centered interactions with AI systems and optimal outcomes in the following ways: • Provides common terminology for describing outcome-based human-AI activities independent of AI techniques and domains • Enables cross-domain insights based on shared human-AI activities • Highlights commonalities in measurement and evaluation needs across disparate AI techniques • Facilitates the development of use cases • Facilitates the evaluation of trustworthiness characteristics and usability.
- Open Access
- 16 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 27-47
Divergences in Blame Attribution after a Security Breach based on Compliance Behavior: Implications for Post-breach Risk Communication
“Attribution of self-blame” is a spontaneous affective and cognitive self-evaluative reaction and is an important predictor of proactive and positive coping response behavior after a negative event. While blame attribution can indeed affect the efficacy of post-breach communication and subsequent behavior in the cybersecurity context, it is not clear whether and how the process of blame attribution may vary in post-data breach contexts based on users’ past security compliance behavior, which can inform the design of post-breach risk communication. As such, to examine the process of blame attribution after a data breach, we run a 2 (user type) x 2 (account type) x 2 (usage scenario) between-group study on Amazon’s MTurk platform. The vignette-based study scenario incorporates multiple stakeholders who may share the responsibility for the data breach (e.g., the negligent user (Bob), the software company holding the data, and the attacker). From the analysis of 255 participant data, we discover that adopters are more likely to hold the protagonist of the story responsible due to his negligent behavior. In contrast, non-adopters are more likely to hold the external entity, such as the service provider, accountable for the account compromise, exhibiting “defensive attribution” of blame. Results further indicate that account types and usage scenarios affect how blame is distributed among different entities. The implications of our findings for effective pre and post-breach risk communication are discussed in the paper.
- 16 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 27-47
Open AccessDivergences in Blame Attribution after a Security Breach based on Compliance Behavior: Implications for Post-breach Risk Communication
“Attribution of self-blame” is a spontaneous affective and cognitive self-evaluative reaction and is an important predictor of proactive and positive coping response behavior after a negative event. While blame attribution can indeed affect the efficacy of post-breach communication and subsequent behavior in the cybersecurity context, it is not clear whether and how the process of blame attribution may vary in post-data breach contexts based on users’ past security compliance behavior, which can inform the design of post-breach risk communication. As such, to examine the process of blame attribution after a data breach, we run a 2 (user type) x 2 (account type) x 2 (usage scenario) between-group study on Amazon’s MTurk platform. The vignette-based study scenario incorporates multiple stakeholders who may share the responsibility for the data breach (e.g., the negligent user (Bob), the software company holding the data, and the attacker). From the analysis of 255 participant data, we discover that adopters are more likely to hold the protagonist of the story responsible due to his negligent behavior. In contrast, non-adopters are more likely to hold the external entity, such as the service provider, accountable for the account compromise, exhibiting “defensive attribution” of blame. Results further indicate that account types and usage scenarios affect how blame is distributed among different entities. The implications of our findings for effective pre and post-breach risk communication are discussed in the paper.
- 25 November 2022
- Published by Springer Nature
- p. 285-306
I’m Only Human: The Effects of Trust Dampening by Anthropomorphic Agents
- 25 November 2022
- Published by Springer Nature
- p. 285-306
I’m Only Human: The Effects of Trust Dampening by Anthropomorphic Agents
- 7 July 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Affective Computing
- Vol. 14 (2) , 1572-1585
The Mediating Effect of Emotions on Trust in the Context of Automated System Usage
Safety-critical systems are often equipped with warning mechanisms to alert users regarding imminent system failures. However, they can suffer from false alarms, and affect users emotions and trust in the system negatively. While providing feedback could be an effective way to calibrate trust under such scenarios, the effects of feedback and warning reliability on users emotions, trust, and compliance behavior is not clear. This paper investigates this by designing a 2 (feedback: present/absent) 2 (warning reliability: high/low) 4 (sessions) mixed design study where participants interacted with a simulated unmanned aerial vehicle (UAV) system to identify and neutralize enemy targets. Results indicated that feedback containing both correctness and affective components decreased users positive emotions and trust in the system, and increased loneliness and hostility (negative) emotions. Emotions were found to mediate the relationship between feedback and trust. Implications of our findings for designing feedback and calibration of trust are discussed in the paper.
- 7 July 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Affective Computing
- Vol. 14 (2) , 1572-1585
The Mediating Effect of Emotions on Trust in the Context of Automated System Usage
Safety-critical systems are often equipped with warning mechanisms to alert users regarding imminent system failures. However, they can suffer from false alarms, and affect users emotions and trust in the system negatively. While providing feedback could be an effective way to calibrate trust under such scenarios, the effects of feedback and warning reliability on users emotions, trust, and compliance behavior is not clear. This paper investigates this by designing a 2 (feedback: present/absent) 2 (warning reliability: high/low) 4 (sessions) mixed design study where participants interacted with a simulated unmanned aerial vehicle (UAV) system to identify and neutralize enemy targets. Results indicated that feedback containing both correctness and affective components decreased users positive emotions and trust in the system, and increased loneliness and hostility (negative) emotions. Emotions were found to mediate the relationship between feedback and trust. Implications of our findings for designing feedback and calibration of trust are discussed in the paper.
- 3 July 2021
- Published by Springer Nature
- p. 41-58
Disentangling Trust and Anthropomorphism Toward the Design of Human-Centered AI Systems
- 3 July 2021
- Published by Springer Nature
- p. 41-58
Disentangling Trust and Anthropomorphism Toward the Design of Human-Centered AI Systems