Pith. sign in

REVIEW 5 minor 62 references

Generative AI Uses and Risks for Knowledge Workers in a Science Organization

T0 review · 0 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read At a US national lab, generative AI use is growing, clustered in copilot and workflow-agent modes, with reliability and security top concerns.

desk verdict A carefully scoped descriptive case study with genuinely new telemetry and qualitative data; limitations are real but acknowledged. read the letter →

arxiv 2501.16577 v1 pith:NRMLFWM7 submitted 2025-01-27 cs.HC cs.AI

Kelly B. Wagman , Matthew T. Dearing , Marshini Chetty This is my paper · ORCID
classification cs.HCcs.AI
keywords generativeAIlargelanguagemodelscopilotworkflowagentknowledgeworkforscienceresponsibleenterprise
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper aims to establish what real-world generative AI use looks like inside a science organization, a setting with both research and operations staff that prior work has not examined together. Using usage logs from an internal AI tool, a survey of 66 employees, and 22 interviews, it argues that early adoption is rising and that current and imagined uses fall into two interaction patterns: copilot interactions, where the AI responds conversationally to user prompts, and workflow agents, where the AI performs tasks with some autonomy. It also contends that employees' main concerns are reliability and hallucinations, overreliance, privacy and security of sensitive data, academic publishing norms, and job impacts. The value of the claim is that it gives organizations a grounded, early-empirical map of where generative AI can help and where it creates risk, informing design and policy decisions.

What carries the argument

The central mechanism is the interaction-modality distinction between a copilot (a conversational system that assists a human on a task) and a workflow agent (a system that performs complex work semi-autonomously). This binary taxonomy organizes the otherwise varied use cases reported by employees and guides the paper's design recommendations. The empirical machinery includes three data sources: telemetry from an organization-specific generative AI interface (a private instance of a commercial LLM), a survey built partly on a prior scientist-focused task list, and semi-structured interviews analyzed with thematic coding. Together these allow the paper to tie usage trends, reported use cases, and concerns to Science and Operations roles.

What would settle it

A lab-wide census survey (not opt-in) of all employees, combined with a full year of usage telemetry, would test the description; if the census found that early-adopter usage rates or concern proportions differed substantially, or if the monthly unique-user trend reversed, the paper's central claims would need revision.

Watch

Extended reading notes

Core claim

The paper's central discovery is a descriptive account of early generative AI adoption in a scientific knowledge-work organization. In the first eight months of an internal AI deployment, monthly unique users grew on average 19.2% (excluding the launch month), with science and operations staff both participating, and survey data showed high familiarity but limited integration into daily workflow. Employees' reported uses sorted neatly into two modalities: copilot-style interactions for writing structured text and code and for envisioned summarization of large unstructured datasets, and workflow-agent applications that automate data pipelines, instrument operations, or project management. The paper also documents that the most common concerns were reliability/hallucinations (44% of survey respondents), privacy/security (42%), overreliance (21%), academic publishing (20%), and job impacts. The authors present these findings as representing early adopters during the first several months of the tool's availability, not the whole organization.

Load-bearing premise

The load-bearing premise is that the 66 survey respondents and 22 interviewees, who were self-selected early adopters, represent the range of generative AI uses and concerns among early adopters across the lab; if that sample is skewed, the reported prevalence numbers will not hold.

Editorial extensions

If this is right

  • Organizations deploying a single internal AI assistant will likely see a slow, experimental uptake initially, with less than 10% of employees using the tool monthly in the first eight months.
  • Design efforts should support both a general-purpose copilot for writing and text summarization and customizable workflow agents for domain-specific automation.
  • The reported concerns imply that organizations need explicit policies for data privacy, AI use in academic publishing, and expectations about job impacts.
  • The observed upward usage trend, if it continues, suggests that generative AI will become a standard part of knowledge work in scientific institutions.
  • The similarity of copilot use across Science and Operations suggests that a single organizational copilot can serve both groups.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the early-adopter pattern generalizes, the 'copilot versus workflow agent' distinction could become a standard way to talk about enterprise generative AI, analogous to the distinction between interactive tools and autonomous agents.
  • The paper's evidence implies that reliability is the main barrier to deeper integration; organizations that can demonstrate high accuracy with citations may see much higher adoption than the early trend suggests.
  • The survey's low job-impact concern (5%) may understate the issue because managers in the study still expected skills to shift; this gap could widen as generative AI capabilities grow.
  • A testable extension would be to track whether employees who start with copilot use later graduate to workflow-agent use, or whether the two modalities remain separate adoption paths.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

0 major / 5 minor

Summary. This paper reports a mixed-methods case study of generative AI use by knowledge workers at a US national laboratory. The authors combine a survey (N=66), semi-structured interviews (N=22), and eight months of telemetry from Argo, an internal GPT-3.5-based chatbot. The reported findings are: (1) Argo use grew during the study period but remained below 10% of lab employees per month; (2) current and envisioned uses group into a 'copilot' modality (structured writing and extracting insights from unstructured text) and a 'workflow agent' modality (automating scientific and operational workflows); (3) Science and Operations employees share many copilot use cases but differ in workflow-agent goals; and (4) participants' main concerns were reliability/hallucinations, overreliance, privacy/security, academic publishing and citation practices, and job impacts. The paper closes with design and policy recommendations for organizations deploying generative AI.

Significance. If the findings hold, this is a useful empirical contribution to a sparsely studied area: it is one of the first organization-level case studies to cover both science and operations roles in a national lab in the first months of an internal LLM deployment. The paper's strengths include transparent reporting of recruitment and coding procedures, a documented two-coder thematic analysis, explicit hedging of the telemetry trend as 'suggestive' (Section 4.1), and a clearly scoped limitation statement (Section 5.3). The copilot/workflow-agent taxonomy is presented as the authors' conceptual framing rather than as a participant-emergent term, which is appropriate. The main value is as a descriptive baseline and design resource; the self-selected, early-adopter sample means prevalence percentages should not be read as population estimates, a caveat the authors mostly acknowledge.

minor comments (5)
  1. [Section 3.2.1] No response rate or denominator for the survey distribution is reported; because Table 3 percentages are presented descriptively, a sentence on the estimated reach of the mailing lists, or an explicit statement that the percentages characterize respondents only, would help readers calibrate them.
  2. [Section 4.2.2] The section is titled 'Envisioned Uses' but includes at least one current use (P8's team used Argo to identify themes in 313 survey responses); consider renaming the section or explicitly separating current from envisioned uses to avoid blurring the paper's current/envisioned distinction.
  3. [Section 4.1 / Figure 1] The 19.2% average monthly increase and the Science/Operations breakdown are computed on small monthly counts, and the figure lacks raw user numbers; adding monthly counts or a small table would make the 'suggestive' trend easier to assess.
  4. [Table 3] The percentages for the privacy/security codes sum to 98.5% (28+12+3+22 out of 66), which suggests either that some responses were not captured or that categories overlap; a footnote explaining how multiply-coded responses were handled and how percentages were rounded would resolve the apparent inconsistency.
  5. [Section 3.2.1] The reason for removing one 'disingenuous' response is not stated; a brief clarification of the exclusion criterion would strengthen transparency.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper's findings are empirical descriptions of telemetry, survey responses, and interviews, with no derivation or fitted parameters that reduce to their own inputs.

full rationale

This paper is an empirical case study reporting descriptive findings from Argo telemetry, a survey (N=66), and interviews (N=22). There is no derivation chain: no equations, no fitted parameters, and no predictive model whose output is defined in terms of its input. The headline usage statistic (19.2% average monthly increase) is computed directly from telemetry and is explicitly hedged as 'suggestive' and 'brief for designating trends' in Section 4.1. The copilot versus workflow-agent taxonomy is introduced by the authors as a conceptual framing, as the paper states in Section 4.2: 'participants themselves rarely used the term copilot, rather we impose it for conceptual organization of the findings.' This is an organizing label applied after data collection, not a result that is predicted from the label, so it is not circular. The survey's task list was drawn from an external study by Morris [30], not from the paper's own conclusions. The only self-citations ([8], [39]) appear in the Background and Discussion as related-work references and are not load-bearing for any finding. The limitations passage in Section 5.3 explicitly notes that participants 'tended to be early adopters of generative AI who were interested in the subject and may not reflect the opinions of employees with little background or interest in the technology'; this is a validity limitation on generalizability, not a circularity in the argument. Overall, the conclusions are empirical descriptions of self-reported and telemetry data, and no step in the paper reduces to its own inputs.

Assumptions & free parameters 0 free parameters · 0 assumptions · 0 invented entities

This is a qualitative empirical study with no equations, fitted parameters, or invented entities. The only assumptions are standard qualitative methodology (self-report validity, coding reliability), which are not formal axioms.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Generative AI Uses and Risks for Knowledge Workers in a Science Organization." pith.science (2026). https://pith.science/paper/NRMLFWM7

@misc{pith2026250116577,
  author       = {Pith},
  title        = {Pith review of: Generative AI Uses and Risks for Knowledge Workers in a Science Organization},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NRMLFWM7}},
  note         = {Machine review of arXiv:2501.16577}
}
read the original abstract

Generative AI could enhance scientific discovery by supporting knowledge workers in science organizations. However, the real-world applications and perceived concerns of generative AI use in these organizations are uncertain. In this paper, we report on a collaborative study with a US national laboratory with employees spanning Science and Operations about their use of generative AI tools. We surveyed 66 employees, interviewed a subset (N=22), and measured early adoption of an internal generative AI interface called Argo lab-wide. We have four findings: (1) Argo usage data shows small but increasing use by Science and Operations employees; Common current and envisioned use cases for generative AI in this context conceptually fall into either a (2) copilot or (3) workflow agent modality; and (4) Concerns include sensitive data security, academic publishing, and job impacts. Based on our findings, we make recommendations for generative AI use in science and other organizations.

Figures

Figures reproduced from arXiv: 2501.16577 by the authors.

Figure 1
Figure 1. Argo usage metrics by Operations and Science unique users for each month since initial deployment. Monthly usage [PITH_FULL_IMAGE:figures/full_fig_p007_1.png] view at source ↗
Figure 2
Figure 2. Responses to the survey question: How familiar are you with large language models (LLMs) such as ChatGPT, Argo, etc.? [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Responses to the survey question: To what extent would you agree/disagree with the following statement: LLMs have become an essential part of my workflow [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Responses to the survey questions: How often do you use LLMs as part of your work? and How often do you use LLMs for personal use? survey respondents said they agreed or strongly agreed that LLMs had become essential to their workflows ( [PITH_FULL_IMAGE:figures/full_…
Figure 5
Figure 5. Figure 5: Responses to the survey question: How often do you use LLMs for the following Argonne National Lab-related work tasks? [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

62 extracted references · 40 canonical work pages

  1. [1]

    Microsoft Research AI4Science and Microsoft Azure Quantum. 2024. The Impact of Large Language Models on Scientific Discovery: a Preliminary Study using GPT-4. https://doi.org/10.48550/arXiv.2311.07361

    arXiv doi:10.48550/arxiv.2311.07361 2024

  2. [2]

    Erik Brynjolfsson, Danielle Li, and Lindsey R. Raymond. 2023. Generative AI at Work. Working Paper 31161. National Bureau of Economic Research. https: //doi.org/10.3386/w31161

    doi:10.3386/w31161 2023

  3. [3]

    Jenna Butler, Sonia Jaffe, Nancy Baym, Mary Czerwinski, Shamsi Iqbal, Kate Nowak, Sean Rintel, Abigail Sellen, Mihaela Vorvoreanu, Brent Hecht, and Jaime Teevan. 2023. Microsoft New Future of Work Report 2023 . Microsoft. https: //aka.ms/nfw2023

    work page 2023

  4. [4]

    Christi Carras. 2024. Major Record Labels Sue AI Companies for Al- legedly Stealing Copyrighted Music. Los Angeles Times (June 2024). https://www.latimes.com/entertainment-arts/business/story/2024-06-24/riaa- suno-udio-lawsuit-ai-copyright-songs-music

    work page 2024

  5. [5]

    Karina Cortiñas-Lorenzo, Siân Lindley, Ida Larsen-Ledet, and Bhaskar Mitra

    work page

  6. [6]

    Mariana Coutinho, Lorena Marques, Anderson Santos, Marcio Dahia, Cesar França, and Ronnie de Souza Santos. 2024. The Role of Generative AI in Software Development Productivity: A Pilot Case Study. In Proceedings of the 1st ACM International Conference on AI-Powered Software (AIware 2024) . Association for Computing Machinery, New York, NY, USA, 131–138. h...

    arXiv 2024

  7. [7]

    Bell, Benjamin Woodward, Henry Ruhl, Kakani Katija, and Angus G

    Alison Crosby, Eric Coughlin Orenstein, Susan E Poulton, Katherine L.C. Bell, Benjamin Woodward, Henry Ruhl, Kakani Katija, and Angus G. Forbes. 2023. Designing Ocean Vision AI: An Investigation of Community Needs for Imaging- based Ocean Conservation. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (CHI ’23) . Association ...

    arXiv 2023

  8. [8]

    Dearing, Yiheng Tao, Xingfu Wu, Zhiling Lan, and Valerie Taylor

    Matthew T. Dearing, Yiheng Tao, Xingfu Wu, Zhiling Lan, and Valerie Taylor

    work page

Show all 62 references
  1. [9]

    Mollick, Hila Lifshitz- Assaf, Katherine Kellogg, Saran Rajendran, Lisa Krayer, François Candelon, and Karim R

    Fabrizio Dell’Acqua, Edward McFowland III, Ethan R. Mollick, Hila Lifshitz- Assaf, Katherine Kellogg, Saran Rajendran, Lisa Krayer, François Candelon, and Karim R. Lakhani. 2023. Navigating the Jagged Technological Frontier: Field Experimental Evidence of the Effects of AI on ...

    2023 doi

  2. [10]

    arXiv:2407.01638 (June 2024)

    LASSI: An LLM-based Automated Self-Correcting Pipeline for Translating Parallel Scientific Codes. arXiv:2407.01638 (June 2024). https://doi.org/10.48550/ arXiv.2407.01638

    2024 doi arXiv

  3. [11]

    It’s like a rubber duck that talks back

    Ian Drosos, Advait Sarkar, Xiaotong Xu, Carina Negreanu, Sean Rintel, and Lev Tankelevitch. 2024. “It’s like a rubber duck that talks back”: Understanding Gen- erative AI-Assisted Data Analysis Workflows through a Participatory Prompting Study. In Proceedings of the 3rd Annual...

    2024

  4. [12]

    Zhuo Diao, Hayato Yamashita, and Masayuki Abe. 2024. Leveraging Large Language Models and Social Media for Automation in Scanning Probe Microscopy. arXiv:2405.15490 (2024). https://doi.org/10.48550/arXiv.2405.15490

    2024 doi arXiv

  5. [13]

    Katy Gero, Alex Calderwood, Charlotte Li, and Lydia Chilton. 2022. A Design Space for Writing Support Tools Using a Cognitive Process Model of Writing. In Proceedings of the First Workshop on Intelligent and Interactive Writing Assistants (In2Writing 2022), Ting-Hao “Kenneth” ...

    2022 doi

  6. [14]

    So what if ChatGPT wrote it?

    Yogesh K. Dwivedi, Nir Kshetri, Laurie Hughes, Emma Louise Slade, Anand Jeyaraj, Arpan Kumar Kar, Abdullah M. Baabdullah, Alex Koohang, Vishnupriya Raghavan, Manju Ahuja, Hanaa Albanna, Mousa Ahmad Albashrawi, Adil S. Al- Busaidi, Janarthanan Balakrishnan, Yves Barlette, Sripa...

    2023

  7. [15]

    William Godoy, Pedro Valero-Lara, Keita Teranishi, Prasanna Balaprakash, and Jeffrey Vetter. 2023. Evaluation of OpenAI Codex for HPC Parallel Programming Models Kernel Generation. In Proceedings of the 52nd International Conference on Parallel Processing Workshops. Associatio...

    2023

  8. [16]

    Mark Glickman and Yi Zhang. 2024. AI and Generative AI for Research Discovery and Summarization. Harvard Data Science Review 6, 2 (March 2024). https: //doi.org/10.1162/99608f92.7f9220ff

    2024 doi

  9. [17]

    Gray and Siddharth Suri

    Mary L. Gray and Siddharth Suri. 2019. Ghost Work: How to Stop Silicon Valley from Building a New Global Underclass . Houghton Mifflin Harcourt, Boston

    2019

  10. [18]

    Gregory Goth. 2023. Why Are Lawyers Afraid of AI? Commun. ACM 67, 1 (Dec. 2023), 14–16. https://doi.org/10.1145/3631935

    2023 doi

  11. [19]

    Perttu Hämäläinen, Mikke Tavast, and Anton Kunnari. 2023. Evaluating Large Language Models in Generating Synthetic HCI Research Data: a Case Study. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (CHI ’23) . Association for Computing Machinery, ...

    2023

  12. [20]

    Melissa Heikkilä. 2023. This New Data Poisoning Tool Lets Artists Fight Back Against Generative AI. MIT Technology Review (2023). https://www.technologyreview.com/2023/10/23/1082189/data-poisoning- artists-fight-generative-ai/

    2023

  13. [21]

    Jean Kaddour, Joshua Harris, Maximilian Mozes, Herbie Bradley, Roberta Raileanu, and Robert McHardy. 2023. Challenges and Applications of Large Language Models. arXiv:2307.10169 (July 2023). https://doi.org/10.48550/arXiv. 2307.10169

    2023 doi arXiv

  14. [22]

    Sonia Jaffe, Neha Parikh Shah, Jenna Butler, Alex Farach, Alexia Cambon, Brent Hecht, Michael Schwarz, and Jaime Teevan (Eds.). 2024. Generative AI in Real- World Workplaces: The Second Microsoft Report on AI and Productivity Research. Microsoft (July 2024). https://www.micros...

    2024

  15. [23]

    Jeongyeon Kim, Sangho Suh, Lydia B Chilton, and Haijun Xia. 2023. Metaphorian: Leveraging Large Language Models to Support Extended Metaphor Creation for Science Writing. In Proceedings of the 2023 ACM Designing Interactive Systems Conference (DIS ’23) . Association for Comput...

    2023

  16. [25]

    Alghamdi, Tal August, Avinash Bhat, Madiha Zahrah Choksi, Senjuti Dutta, Jin L.C

    Mina Lee, Katy Ilonka Gero, John Joon Young Chung, Simon Buckingham Shum, Vipul Raheja, Hua Shen, Subhashini Venugopalan, Thiemo Wambsganss, David Zhou, Emad A. Alghamdi, Tal August, Avinash Bhat, Madiha Zahrah Choksi, Senjuti Dutta, Jin L.C. Guo, Md Naimul Hoque, Yewon Kim, S...

  17. [26]

    If the Machine Is As Good As Me, Then What Use Am I?

    Charlotte Kobiella, Yarhy Said Flores López, Franz Waltenberger, Fiona Draxler, and Albrecht Schmidt. 2024. “If the Machine Is As Good As Me, Then What Use Am I?” – How the Use of ChatGPT Changes Young Professionals’ Perception of Productivity and Accomplishment. InProceedings...

    2024

  18. [27]

    Steve Lohr. 2024. Generative A.I. ’s Biggest Impact Will Be in Banking and Tech, Report Says. The New York Times (Feb. 2024). https://www.nytimes.com/2024/ 02/01/business/ai-impact-jobs.html

    2024

  19. [28]

    In Proceedings of the CHI Conference on Human Factors in Computing Systems (CHI ’24)

    A Design Space for Intelligent and Interactive Writing Assistants. In Proceedings of the CHI Conference on Human Factors in Computing Systems (CHI ’24) . Association for Computing Machinery, New York, NY, USA, 1–35. https://doi.org/10.1145/3613904.3642697

  20. [29]

    Ge Lei, Ronan Docherty, and Samuel J. Cooper. 2024. Materials science in the era of large language models: a perspective. Digital Discovery. Digital Discovery 3, 7 (2024), 1257–1272. https://doi.org/10.1039/D4DD00074A

    2024 doi

  21. [30]

    Meredith Ringel Morris. 2023. Scientists’ Perspectives on the Potential for Gener- ative AI in their Fields. arXiv:2304.01420 (April 2023). https://doi.org/10.48550/ arXiv.2304.01420

    2023 doi arXiv

  22. [31]

    Angus Loten. 2023. Generative AI Tools Use Custom Data to Power More Business Functions. Wall Street Journal (March 2023). https://www.wsj.com/articles/generative-ai-tools-use-custom-data-to- power-more-business-functions-2f158dbb

    2023

  23. [32]

    Paula Maddigan and Teo Susnjak. 2023. Chat2VIS: Generating Data Visualiza- tions via Natural Language Using ChatGPT, Codex and GPT-3 Large Language Models. IEEE Access 11 (2023), 45181–45193. https://doi.org/10.1109/ACCESS. 2023.3274199 Generative AI Uses and Risks for Knowled...

    2023

  24. [34]

    Davis, Zhaojun Xie, Arjun Rajaram, and Abhinav Bhatele

    Daniel Nichols, Joshua H. Davis, Zhaojun Xie, Arjun Rajaram, and Abhinav Bhatele. 2024. Can Large Language Models Write Parallel Code?. In Proceedings of the 33rd International Symposium on High-Performance Parallel and Distributed Computing. Association for Computing Machiner...

    2024

  25. [35]

    Shakked Noy and Whitney Zhang. 2023. Experimental evidence on the produc- tivity effects of generative artificial intelligence. Science 381, 6654 (July 2023), 187–192. https://doi.org/10.1126/science.adh2586

    2023 doi

  26. [36]

    Sida Peng, Eirini Kalliamvakou, Peter Cihon, and Mert Demirer. 2023. The Impact of AI on Developer Productivity: Evidence from GitHub Copilot. arXiv:2302.06590 (Feb. 2023). https://doi.org/10.48550/arXiv.2302.06590

    2023 doi arXiv

  27. [37]

    Samir Passi and Mihaela Vorvoreanu. 2022. Overreliance on AI: Literature Review . Microsoft Technical Report MSR-TR-2022-12. Microsoft. https://www.microsoft. com/en-us/research/publication/overreliance-on-ai-literature-review/

    2022

  28. [38]

    Hammond Pearce, Baleegh Ahmad, Benjamin Tan, Brendan Dolan-Gavitt, and Ramesh Karri. 2022. Asleep at the Keyboard? Assessing the Security of GitHub Copilot’s Code Contributions. In 2022 IEEE Symposium on Security and Privacy (SP). 754–768. https://doi.org/10.1109/SP46214.2022.9833571

    2022

  29. [39]

    Michael H Prince, Henry Chan, Aikaterini Vriza, Tao Zhou, Varuni K Sastry, Matthew T Dearing, Ross J Harder, Rama K Vasudevan, and Mathew J Cherukara

  30. [40]

    Neil Perry, Megha Srivastava, Deepak Kumar, and Dan Boneh. 2023. Do Users Write More Insecure Code with AI Assistants?. In Proceedings of the 2023 ACM SIGSAC Conference on Computer and Communications Security . 2785–2799. https: //doi.org/10.1145/3576915.3623157

    2023

  31. [41]

    Savvas Petridis, Nicholas Diakopoulos, Kevin Crowston, Mark Hansen, Keren Henderson, Stan Jastrzebski, Jeffrey V Nickerson, and Lydia B Chilton. 2023. AngleKindling: Supporting Journalistic Angle Ideation with Large Language Models. In Proceedings of the 2023 CHI Conference on...

    2023

  32. [42]

    Daniel Russo. 2024. Navigating the Complexity of Generative AI Adoption in Software Engineering. ACM Trans. Softw. Eng. Methodol. 33, 5 (June 2024), 135:1–135:50. https://doi.org/10.1145/3652154

    2024 doi

  33. [43]

    Mari Sako. 2024. How Generative AI Fits into Knowledge Work. Commun. ACM 67, 4 (March 2024), 20–22. https://doi.org/10.1145/3638567

    2024 doi

  34. [44]

    Francesco Regazzoni, Paolo Palmieri, Fethulah Smailbegovic, Rosario Cammarota, and Ilia Polian. 2021. Protecting artificial intelligence IPs: a survey of watermark- ing and fingerprinting for machine learning. CAAI Transactions on Intelligence Technology 6, 2 (2021), 180–191. ...

    2021 doi

  35. [45]

    Alex Reisner. 2023. These 183,000 Books Are Fueling the Biggest Fight in Publishing and Tech. The Atlantic (Sept. 2023). https: //www.theatlantic.com/technology/archive/2023/09/books3-database- generative-ai-training-copyright-infringement/675363/

    2023

  36. [46]

    Abigail Sellen and Eric Horvitz. 2024. The Rise of the AI Co-Pilot: Lessons for Design from Aviation and Beyond. Commun. ACM 67, 7 (July 2024), 18–23. https://doi.org/10.1145/3637865

    2024 doi

  37. [47]

    Shawn Shan, Wenxin Ding, Josephine Passananti, Stanley Wu, Haitao Zheng, and Ben Y. Zhao. 2024. Nightshade: Prompt-Specific Poisoning Attacks on Text-to-Image Generative Models. IEEE Computer Society, 807–825. https: //doi.org/10.1109/SP54263.2024.00207

    2024

  38. [48]

    Johnny Saldañna. 2013. The Coding Manual for Qualitative Researchers (2nd ed ed.). SAGE, Los Angeles

    2013

  39. [49]

    Timo Schick, Jane Dwivedi-Yu, Roberto Dessì, Roberta Raileanu, Maria Lomeli, Eric Hambro, Luke Zettlemoyer, Nicola Cancedda, and Thomas Scialom. 2023. Toolformer: Language Models Can Teach Themselves to Use Tools. InAdvances in Neural Information Processing Systems (NeurIPS ’2...

    2023

  40. [50]

    Mann, Michael Irvin, J

    Shuaiwen Leon Song, Bonnie Kruft, Minjia Zhang, Conglong Li, Shiyang Chen, Chengming Zhang, Masahiro Tanaka, Xiaoxia Wu, Jeff Rasley, Ammar Ahmad Awan, Connor Holmes, Martin Cai, Adam Ghanem, Zhongzhu Zhou, Yuxiong He, Pete Luferenko, Divya Kumar, Jonathan Weyn, Ruixiong Zhang...

    2023 doi arXiv

  41. [51]

    Minhyang (Mia) Suh, Emily Youngblom, Michael Terry, and Carrie J Cai. 2021. AI as Social Glue: Uncovering the Roles of Deep Generative AI during Social Music Composition. In Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems (CHI ’21) . Association fo...

    2021

  42. [52]

    Jingyu Shi, Rahul Jain, Hyungjun Doh, Ryo Suzuki, and Karthik Ramani. 2024. An HCI-Centric Survey and Taxonomy of Human-Generative-AI Interactions. arXiv:2310.07127 (Jan. 2024). https://doi.org/10.48550/arXiv.2310.07127

    2024 doi arXiv

  43. [53]

    Delavaux, Camille Fournier de Lauriere, Johan van den Hoogen, Thomas Lauber, Haozhi Ma, Daniel S

    Gabriel Reuben Smith, Carolina Bello, Lalasia Bialic-Murphy, Emily Clark, Camille S. Delavaux, Camille Fournier de Lauriere, Johan van den Hoogen, Thomas Lauber, Haozhi Ma, Daniel S. Maynard, Matthew Mirman, Lidong Mo, Do- minic Rebindaine, Josephine Elena Reek, Leland K. Werd...

    2024

  44. [54]

    Mario Sänger, Ninon De Mecquenem, Katarzyna Ewa Lewińska, Vasilis Bountris, Fabian Lehmann, Ulf Leser, and Thomas Kosch. 2024. A Qualitative Assessment of Using ChatGPT as Large Language Model for Scientific Workflow Development. GigaScience 13 (2024), 1–19. https://doi.org/10...

    2024 doi

  45. [55]

    Ross Taylor, Marcin Kardas, Guillem Cucurull, Thomas Scialom, Anthony Hartshorn, Elvis Saravia, Andrew Poulton, Viktor Kerkez, and Robert Stojnic

  46. [56]

    Sangho Suh, Bryan Min, Srishti Palani, and Haijun Xia. 2023. Sensecape: En- abling Multilevel Exploration and Sensemaking with Large Language Models. In Proceedings of the 36th Annual ACM Symposium on User Interface Software and Technology (UIST ’23) . Association for Computin...

    2023

  47. [57]

    Yuan Sun, Eunchae Jang, Fenglong Ma, and Ting Wang. 2024. Generative AI in the Wild: Prospects, Challenges, and Strategies. In Proceedings of the CHI Conference on Human Factors in Computing Systems (CHI ’24) . Association for Computing Machinery, New York, NY, USA, 1–16. http...

    2024

  48. [58]

    Ruotong Wang, Ruijia Cheng, Denae Ford, and Thomas Zimmermann. 2024. Investigating and Designing for Trust in AI-powered Code Generation Tools. In Proceedings of the 2024 ACM Conference on Fairness, Accountability, and Trans- parency (FAccT ’24). Association for Computing Mach...

    2024

  49. [59]

    Allison Woodruff, Renee Shelby, Patrick Gage Kelley, Steven Rousso-Schindler, Jamila Smith-Loud, and Lauren Wilcox. 2024. How Knowledge Workers Think Generative AI Will (Not) Transform Their Industries. In Proceedings of the CHI Conference on Human Factors in Computing Systems...

    2024

  50. [61]

    Clay to Play With

    Severi Uusitalo, Antti Salovaara, Tero Jokela, and Marja Salmimaa. 2024. “Clay to Play With”: Generative AI Tools in UX and Industrial Design Practice. In Proceedings of the 2024 ACM Designing Interactive Systems Conference (DIS ’24) . Association for Computing Machinery, New ...

    2024

  51. [62]

    Jiyao Wang, Chunxi Huang, Song Yan, Weiyin Xie, and Dengbo He. 2024. When Young Scholars Cooperate with LLMs in Academic Tasks: The Influence of Individ- ual Differences and Task Complexities.International Journal of Human–Computer Interaction (2024), 1–16. https://doi.org/10....

    2024

  52. [2022]

    arXiv:2211.09085 (Nov

    Galactica: A Large Language Model for Science. arXiv:2211.09085 (Nov. 2022). https://doi.org/10.48550/arXiv.2211.09085

    2022 doi arXiv

  53. [2023]

    arXiv:2312.01291 (2023)

    Opportunities for Retrieval and Tool Augmented Large Language Models in Scientific Facilities. arXiv:2312.01291 (2023). https://doi.org/10.48550/arXiv. 2312.01291

    2023 doi arXiv

  54. [2024]

    arXiv:2401.09410 (Jan

    Through the Looking-Glass: Transparency Implications and Challenges in Enterprise AI Knowledge Systems. arXiv:2401.09410 (Jan. 2024). https: //doi.org/10.48550/arXiv.2401.09410

    2024 doi arXiv

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.