pith. sign in

arxiv: 2606.18716 · v1 · pith:SV5KHJWUnew · submitted 2026-06-17 · 💻 cs.HC · cs.AI

Human-AI Agent Interaction in a Business Context

Kathrin Paimann , Elizangela Valarini , Sebastian Juhl This is my paper

Pith reviewed 2026-06-26 19:53 UTC · model grok-4.3

classification 💻 cs.HC cs.AI
keywords human-AI interactionAI agentsuser experiencebusiness contextmixed methodsadoptiontrust
0
0 comments X

The pith

Identifying user expectations for AI agents in business facilitates adoption and builds trust.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper uses a mixed-methods approach to explore interaction patterns between humans and AI agents in business contexts. It seeks to identify principles and criteria for positive user experience along with ways to measure it. The goal is to understand user needs so that development teams can make user-centered decisions. This exploratory work lays the foundation for a larger survey experiment to test the effectiveness of design elements.

Core claim

Through qualitative and quantitative techniques, the study identifies user expectations and needs that, when addressed in design, facilitate adoption, build trust, and support better decision-making by teams developing AI agents for business processes.

What carries the argument

Mixed-methods exploration of human-AI agent interaction patterns to derive UX principles and measurement methods.

If this is right

  • Principles and criteria for positive UX with AI agents can be established.
  • Methods for measuring UX in these interactions can be developed and evaluated.
  • Development teams gain support for user-centered decision-making.
  • Adoption of AI agents in business is facilitated through better alignment with user needs.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • These principles could extend to non-business applications if similar user needs are present.
  • A follow-up large-scale survey might identify which design elements have the strongest impact on trust.
  • Teams could use the identified criteria to create standardized evaluation frameworks for AI agents.

Load-bearing premise

The mixed-methods exploration will produce principles and criteria that remain effective when evaluated through a larger-scale survey experiment.

What would settle it

A larger-scale survey experiment that shows the identified design elements have no positive effect on user experience metrics would falsify the claim.

Figures

Figures reproduced from arXiv: 2606.18716 by Elizangela Valarini, Kathrin Paimann, Sebastian Juhl.

Figure 1
Figure 1. Figure 1: Average Marginal Component Effects of the UX criteria related to the human control UX principle. • Ability to stop/pause agents: Participants consistently emphasized the need to interrupt agent execution, reflecting a fundamental re￾quirement for human authority over automated processes. • Transparency of agent reasoning: Participants required visibility into agent reasoning and actions. This criterion is … view at source ↗
read the original abstract

As AI agents are increasingly integrated into core business processes, understanding and designing effective interaction patterns between humans and AI agents becomes crucial for value creation. This study identifies and evaluates principles and criteria for a positive User Experience (UX) with AI agents, along with methods for its measurement. We identify user expectations and needs to facilitate adoption, build trust, and support user-centered decision-making by development teams. Using a mixed-methods approach that combines qualitative and quantitative techniques, we explore interaction patterns between humans and AI agents. The findings from this exploratory research serve as the basis to develop a survey experiment which evaluates the effectiveness of specific design elements on a larger scale. This foundational research contributes to the development of more intuitive and effective human-AI agent interactions in business settings.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript outlines an exploratory mixed-methods study to identify principles and criteria for positive UX with AI agents in business contexts, with the goal of identifying user expectations to facilitate adoption, build trust, and support user-centered decision-making by development teams. Findings from this work are intended to inform a subsequent larger-scale survey experiment. No data, results, specific principles, or evaluations are reported.

Significance. The topic of human-AI agent interaction is relevant to HCI. However, because the manuscript presents only a high-level research plan without any empirical findings, validated principles, or tested claims, its significance is limited to the potential of the planned work rather than any demonstrated contribution.

major comments (2)
  1. [Abstract] Abstract: The manuscript states that the study 'identifies and evaluates principles and criteria for a positive User Experience (UX) with AI agents' but provides no principles, criteria, data, or results, leaving the central contribution without empirical support.
  2. [Abstract] Abstract: The assertion that identifying user expectations 'facilitates adoption, builds trust, and supports user-centered decision-making' is presented as a premise rather than a tested finding, with no evidence or validation offered in the described study design.
minor comments (1)
  1. The mixed-methods approach is described only at a high level; additional detail on specific qualitative and quantitative techniques would improve the clarity of the research plan.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed review. The manuscript presents the design of an exploratory mixed-methods study rather than completed empirical results, and we will revise the abstract to ensure the language accurately describes the planned work and its intended goals without implying completed findings.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The manuscript states that the study 'identifies and evaluates principles and criteria for a positive User Experience (UX) with AI agents' but provides no principles, criteria, data, or results, leaving the central contribution without empirical support.

    Authors: We agree the abstract phrasing implies completed identification and evaluation. The manuscript details the research design for a mixed-methods exploratory phase (qualitative exploration of interaction patterns to surface principles, followed by a planned larger survey experiment) but reports no data or validated principles because the work is at the design stage. We will revise the abstract to state that the study outlines methods to identify and evaluate such principles, with findings intended to inform subsequent validation. revision: yes

  2. Referee: [Abstract] Abstract: The assertion that identifying user expectations 'facilitates adoption, builds trust, and supports user-centered decision-making' is presented as a premise rather than a tested finding, with no evidence or validation offered in the described study design.

    Authors: The phrasing reflects the motivating rationale drawn from existing HCI literature on AI agent adoption. We acknowledge it should not be framed as a tested outcome of the current study. We will revise the abstract to present these as the anticipated benefits and goals that the exploratory research aims to enable, rather than as demonstrated results. revision: yes

Circularity Check

0 steps flagged

No significant circularity; purely exploratory plan with no derivations

full rationale

The manuscript describes a planned mixed-methods exploration whose stated purpose is to identify UX principles for later evaluation in a survey experiment. No equations, fitted parameters, uniqueness theorems, or self-citations appear in the provided text. The central statement that identifying expectations facilitates adoption is presented as a motivating premise rather than a result derived from any internal construction or prior author work. Because the paper contains no completed data, no predictive claims, and no load-bearing steps that reduce to their own inputs, the work is self-contained as descriptive research and receives a score of zero.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review shows no mathematical models, free parameters, axioms, or invented entities; the work is qualitative and exploratory without formal modeling.

pith-pipeline@v0.9.1-grok · 5653 in / 974 out tokens · 31991 ms · 2026-06-26T19:53:49.734546+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

59 extracted references · 30 canonical work pages

  1. [1]

    Political Analysis ,author=

    Causal Inference in Conjoint Analysis: Understanding Multidimensional Choices via Stated Preference Experiments ,volume=. Political Analysis ,author=. 2014 ,pages=. doi:10.1093/pan/mpt024 ,number=

    work page doi:10.1093/pan/mpt024 2014

  2. [2]

    Political Analysis ,author=

    Multiple Hypothesis Testing in Conjoint Analysis ,volume=. Political Analysis ,author=. 2023 ,pages=. doi:10.1017/pan.2022.30 ,number=

    work page doi:10.1017/pan.2022.30 2023

  3. [3]

    Political Analysis ,author=

    The Number of Choice Tasks and Survey Satisficing in Conjoint Experiments ,volume=. Political Analysis ,author=. 2018 ,pages=. doi:10.1017/pan.2017.40 ,number=

    work page doi:10.1017/pan.2017.40 2018

  4. [4]

    Political Science Research and Methods ,author=

    Assessing the relative influence of party unity on vote choice: evidence from a conjoint experiment ,volume=. Political Science Research and Methods ,author=. 2024 ,pages=. doi:10.1017/psrm.2022.45 ,number=

    work page doi:10.1017/psrm.2022.45 2024

  5. [5]

    Political Analysis ,author=

    Using Eye-Tracking to Understand Decision-Making in Conjoint Experiments ,volume=. Political Analysis ,author=. 2021 ,pages=. doi:10.1017/pan.2020.11 ,number=

    work page doi:10.1017/pan.2020.11 2021

  6. [6]

    Green and Vithala R

    Paul E. Green and Vithala R. Rao , journal =. Conjoint Measurement for Quantifying Judgmental Data , volume =

  7. [7]

    Political Analysis ,author=

    Using Conjoint Experiments to Analyze Election Outcomes: The Essential Role of the Average Marginal Component Effect ,volume=. Political Analysis ,author=. 2023 ,pages=. doi:10.1017/pan.2022.16 ,number=

    work page doi:10.1017/pan.2022.16 2023

  8. [8]

    and Bottoni, Paolo and Pareschi, Remo , title=

    Borghoff, Uwe M. and Bottoni, Paolo and Pareschi, Remo , title=. Frontiers in Human Dynamics , volume=. 2025 , DOI=

    2025

  9. [9]

    2026 , issn =

    Agentic Business Process Management: A research manifesto , journal =. 2026 , issn =. doi:https://doi.org/10.1016/j.is.2026.102738 , author =

    work page doi:10.1016/j.is.2026.102738 2026

  10. [10]

    2026 , isbn =

    Cheng, Ruijia and Liang, Jenny T and Schoop, Eldon and Nichols, Jeffrey , title =. 2026 , isbn =. doi:10.1145/3742413.3789132 , pages =

    work page doi:10.1145/3742413.3789132 2026

  11. [11]

    2022 , author =

    On the Design of and Interaction with Conversational Agents: An Organizing and Assessing Review of Human-Computer Interaction Research , journal =. 2022 , author =

    2022

  12. [12]

    2025 , issn =

    Human-AI agency in the age of generative AI , journal =. 2025 , issn =. doi:https://doi.org/10.1016/j.infoandorg.2025.100560 , author =

    work page doi:10.1016/j.infoandorg.2025.100560 2025

  13. [13]

    Journal of Marketing , volume =

    From Tools to Agents: Meta-Analytic Insights into Human Acceptance of AI , author =. Journal of Marketing , volume =. 2026 , doi =

    2026

  14. [14]

    2025 , note =

    Microsoft Design , title =. 2025 , note =

    2025

  15. [15]

    2026 , eprint=

    A User Experience 3.0 (UX 3.0) Paradigm Framework: Designing for Human-Centered AI Experiences , author=. 2026 , eprint=

    2026

  16. [16]

    User Experience 3.0 (UX 3.0)

    A "User Experience 3.0 (UX 3.0)" Paradigm Framework: User Experience Design for Human-Centered AI Systems , author=. 2024 , eprint=

    2024

  17. [17]

    The American Journal of Bioethics , volume =

    Elise Li Zheng and Weina Jin and Ghassan Hamarneh and Sandra Soo-Jin Lee , title =. The American Journal of Bioethics , volume =. 2024 , publisher =

    2024

  18. [18]

    and Feltovich, Paul and Johnson, Matthew , title =

    Bradshaw, Jeffrey M. and Feltovich, Paul and Johnson, Matthew , title =. Handbook of Human-Machine Interaction , editor =. 2011 , chapter =

    2011

  19. [19]

    2023 , url =

    Newman, Jessica , title =. 2023 , url =

    2023

  20. [20]

    2023 , doi =

    Artificial Intelligence Risk Management Framework (AI RMF 1.0) , institution =. 2023 , doi =

    2023

  21. [21]

    Learning paradigms for hybrid decision-making systems.ACM Comput

    Punzi, Clara and Pellungrini, Roberto and Setzu, Mattia and Giannotti, Fosca and Pedreschi, Dino , title =. 2026 , publisher =. doi:10.1145/3802522 , journal =

    work page doi:10.1145/3802522 2026

  22. [22]

    International Journal of Human–Computer Interaction , volume =

    Ben Shneiderman , title =. International Journal of Human–Computer Interaction , volume =. 2020 , publisher =

    2020

  23. [23]

    2025 , eprint=

    An HCAI Methodological Framework (HCAI-MF): Putting It Into Action to Enable Human-Centered AI , author=. 2025 , eprint=

    2025

  24. [24]

    Journal of Participatory Medicine , volume =

    Arcia, Adriana and Stonbraker, Samantha and Mangal, Sabrina and Lor, Maichou , title =. Journal of Participatory Medicine , volume =. 2024 , doi =

    2024

  25. [25]

    1991 , edition =

    Design at Work: Cooperative Design of Computer Systems , publisher =. 1991 , edition =. doi:10.1201/9781003063988 , url =

    work page doi:10.1201/9781003063988 1991

  26. [26]

    ocking, Mara and Ose, Claudia and Znotka, Matti and C\

    Minartz, Peter and Vondeberg, Carmen and Obert, Laura and Bleck, Christian and Hemming, Bernhard and V\"ocking, Mara and Ose, Claudia and Znotka, Matti and C\"urlis, Diana and Karrenbrock, Anne and Oehmichen, Frank and Heistermann, Peter and Kuske, Silke , title =. Das Gesundheitswesen , year =

  27. [27]

    , title =

    Muller, Michael J. , title =. The Human-Computer Interaction Handbook: Fundamentals, Evolving Technologies and Emerging Applications , editor =. 2008 , isbn =

    2008

  28. [28]

    and Kuhn, Sarah , title =

    Muller, Michael J. and Kuhn, Sarah , title =. 1993 , issue_date =. doi:10.1145/153571.255960 , journal =

    work page doi:10.1145/153571.255960 1993

  29. [29]

    Co-Designing a Web-Based Decision Aid Tool for Employees Disclosure of Mental Health Conditions: A Participatory Study Design Using Employee and Organizational Preferences

    Stratton, Elizabeth and Choi, Isabella and Peters, Dorian and Calvo, Rafael A and Harvey, Samuel B and Glozier, Nicholas. Co-Designing a Web-Based Decision Aid Tool for Employees Disclosure of Mental Health Conditions: A Participatory Study Design Using Employee and Organizational Preferences. JMIR Formative Research. 2020. doi:10.2196/23337

    work page doi:10.2196/23337 2020

  30. [30]

    1993 , doi =

    Schuler, Douglas and Namioka, Aki , title =. 1993 , doi =

    1993

  31. [31]

    and Verma, Aditi ,year=

    Wacnik, Peter and Daly, Shanna R. and Verma, Aditi ,year=. Participatory design: a systematic review and insights for future practice ,volume=. doi:10.1017/dsj.2025.10009 ,journal=

    work page doi:10.1017/dsj.2025.10009 2025

  32. [32]

    BMC Health Services Research , year =

    Sattar, Raabia and Lawton, Rebecca and Panagioti, Maria and Johnson, Judith , title =. BMC Health Services Research , year =. doi:10.1186/s12913-020-06049-w , url =

    work page doi:10.1186/s12913-020-06049-w

  33. [33]

    Psychology & Marketing , volume =

    Paul, Justin and Barari, Mojtaba , title =. Psychology & Marketing , volume =. doi:https://doi.org/10.1002/mar.21657 , year =

    work page doi:10.1002/mar.21657

  34. [34]

    2022 , isbn =

    Ralph, Paul and Baltes, Sebastian , title =. 2022 , isbn =. doi:10.1145/3540250.3560877 , booktitle =

    work page doi:10.1145/3540250.3560877 2022

  35. [35]

    Frontiers in Psychiatry , VOLUME=

    Lachal, Jonathan and Revah-Levy, Anne and Orri, Massimiliano and Moro, Marie Rose , TITLE=. Frontiers in Psychiatry , VOLUME=. 2017 , URL=. doi:10.3389/fpsyt.2017.00269 , ISSN=

    work page doi:10.3389/fpsyt.2017.00269 2017

  36. [36]

    2007 , publisher=

    Handbook for Synthesizing Qualitative Research , author=. 2007 , publisher=

    2007

  37. [37]

    2024 , issn =

    Ecosystem orchestration practices for industrial firms: A qualitative meta-analysis, framework development and research agenda , journal =. 2024 , issn =. doi:https://doi.org/10.1016/j.jbusres.2023.114463 , author =

    work page doi:10.1016/j.jbusres.2023.114463 2024

  38. [38]

    2004 , doi =

    Boynton, Petra M and Greenhalgh, Trisha , title =. 2004 , doi =. https://www.bmj.com/content/328/7451/1312.full.pdf , journal =

    2004

  39. [39]

    and Duffett, Mark and Kho, Michelle E

    Burns, Karen E.A. and Duffett, Mark and Kho, Michelle E. and Meade, Maureen O. and Adhikari, Neill K.J. and Sinuff, Tasnim and Cook, Deborah J. , title =. 2008 , doi =

    2008

  40. [40]

    2023 , doi =

    Ranganathan, Priya and Caduff, Carlo , title =. 2023 , doi =

    2023

  41. [41]

    PLOS ONE , publisher =

    Paper-and-pencil versus computerized administration mode: Comparison of data quality and risk behavior prevalence estimates in the European school Survey Project on Alcohol and other Drugs (ESPAD) , year =. PLOS ONE , publisher =. doi:10.1371/journal.pone.0225140 , author =

    work page doi:10.1371/journal.pone.0225140

  42. [42]

    BMC Medical Research Methodology , year =

    Edwards, Phil and Perkins, Chloe , title =. BMC Medical Research Methodology , year =. doi:10.1186/s12874-024-02332-0 , url =

    work page doi:10.1186/s12874-024-02332-0

  43. [43]

    Braun and V

    Virginia Braun and Victoria Clarke , title =. Qualitative Research in Psychology , volume =. 2006 , publisher =. doi:10.1191/1478088706qp063oa , URL =

    work page doi:10.1191/1478088706qp063oa 2006

  44. [44]

    Qualitative Research in Sport, Exercise and Health , volume =

    Virginia Braun and Victoria Clarke , title =. Qualitative Research in Sport, Exercise and Health , volume =. 2019 , publisher =. doi:10.1080/2159676X.2019.1628806 , URL =

    work page doi:10.1080/2159676x.2019.1628806 2019

  45. [45]

    , title =

    DeJonckheere, Melissa and Vaughn, Lisa M. , title =. Family Medicine and Community Health , year =. doi:10.1136/fmch-2018-000057 , url =

    work page doi:10.1136/fmch-2018-000057 2018

  46. [46]

    and Coates, Wendy C

    Jordan, Jaime and Clarke, Samuel O. and Coates, Wendy C. , title =. AEM Education and Training , volume =. doi:https://doi.org/10.1002/aet2.10646 , url =. https://onlinelibrary.wiley.com/doi/pdf/10.1002/aet2.10646 , year =

    work page doi:10.1002/aet2.10646

  47. [47]

    1996 , publisher =

    Kvale, Steinar , title =. 1996 , publisher =

    1996

  48. [48]

    McIntosh and Janice M

    Michele J. McIntosh and Janice M. Morse , title =. Global Qualitative Nursing Research , volume =. 2015 , doi =

    2015

  49. [49]

    2020 , edition =

    Russell, Stuart and Norvig, Peter , title =. 2020 , edition =

    2020

  50. [50]

    2023 , issn =

    Sample size selection for discrete choice experiments using design features , journal =. 2023 , issn =. doi:https://doi.org/10.1016/j.jocm.2023.100436 , url =

    work page doi:10.1016/j.jocm.2023.100436 2023

  51. [51]

    Sawtooth Software Research Paper Series , year =

    Johnson, Richard and Orme, Bryan , title =. Sawtooth Software Research Paper Series , year =

  52. [52]

    2019 , publisher =

    Orme, Bryan , title =. 2019 , publisher =

    2019

  53. [53]

    Cai, Emily Reif, Narayan Hegde, Jason Hipp, Been Kim, Daniel Smilkov, Martin Wattenberg, Fernanda Viegas, Greg S

    Amershi, Saleema and Weld, Dan and Vorvoreanu, Mihaela and Fourney, Adam and Nushi, Besmira and Collisson, Penny and Suh, Jina and Iqbal, Shamsi and Bennett, Paul N. and Inkpen, Kori and Teevan, Jaime and Kikin-Gil, Ruth and Horvitz, Eric , title =. 2019 , isbn =. doi:10.1145/3290605.3300233 , booktitle =

    work page doi:10.1145/3290605.3300233 2019

  54. [54]

    Electronics , VOLUME =

    Nastoska, Aleksandra and Jancheska, Bojana and Rizinski, Maryan and Trajanov, Dimitar , TITLE =. Electronics , VOLUME =. 2025 , NUMBER =

    2025

  55. [55]

    Notre Dame Law Review , volume =

    Kolt, Noam , title =. Notre Dame Law Review , volume =. 2025 , note =

    2025

  56. [56]

    Journal of Association for Information Systems , volume =

    Zhang, Ping and Li, Na , title =. Journal of Association for Information Systems , volume =. 2005 , pages =. doi:10.17705/1jais.00070 , url =

    work page doi:10.17705/1jais.00070 2005

  57. [57]

    Future Internet , VOLUME =

    Bandi, Ajay and Kongari, Bhavani and Naguru, Roshini and Pasnoor, Sahitya and Vilipala, Sri Vidya , TITLE =. Future Internet , VOLUME =. 2025 , NUMBER =. doi:10.3390/fi17090404 , PAGES =

    work page doi:10.3390/fi17090404 2025

  58. [58]

    and Keever, David , booktitle=

    Quinn, Mary and Piper, Blake and Bliss, James P. and Keever, David , booktitle=. Recommended Methods for Using the 2020 NIST Principles for AI Explainability , year=

    2020

  59. [59]

    2025 , issue_date =

    Deng, Zehang and Guo, Yongjian and Han, Changzhou and Ma, Wanlun and Xiong, Junwu and Wen, Sheng and Xiang, Yang , title =. 2025 , issue_date =. doi:10.1145/3716628 , journal =

    work page doi:10.1145/3716628 2025