@unpublished{CardonGetchellCarradinietal.2023, author = {Cardon, Peter W. and Getchell, Kristen and Carradini, Stephen and Fleischmann, Carolin and Stapp, James}, title = {Generative AI in the Workplace: Employee Perspectives of ChatGPT Benefits and Organizational Policies}, doi = {10.31235/osf.io/b3ezy}, year = {2023}, abstract = {Research Questions • What attitudes do professionals hold towards the impact of AI on society and their jobs? How do early adopters of ChatGPT differ in their attitudes towards the impact of AI on society and their jobs? • In what ways are professionals using ChatGPT? How do these uses differ by managerial status? • What do professional perceive as benefits of generative AI? How do they differ in these views based on level of ChatGPT adoption and managerial status? • What do professionals perceive as the benefits of organizational policy for the use of generative AI? How do these views differ based on level of ChatGPT adoption and managerial status? Samples • Study 1 involved 148 working adults in the United States who had heard of ChatGPT • Study 2 involved 395 working adults in the United States who had heard of ChatGPT Research Questions • What attitudes do professionals hold towards the impact of AI on society and their jobs? How do early adopters of ChatGPT differ from non-adopters in their attitudes towards the impact of AI on society and their jobs? • In what ways are professionals using ChatGPT? How do these uses differ by managerial status? • What do professionals perceive as benefits of generative AI? How do they differ in these views based on level of ChatGPT adoption and managerial status? • What do professionals perceive as the benefits of organizational policy for the use of generative AI? How do these views differ based on level of ChatGPT adoption and managerial status? Key Findings and Conclusions • Many US workers in this sample are using ChatGPT for professional purposes. Roughly the following percentages have already used ChatGPT in the following ways: o 42\% for researching a topic or generating ideas o 32\% for drafting messages o 26\% for drafting longer documents, such as reports o 22\% for editing text • Many US workers in this sample believe ChatGPT can help them become better communicators. This is particularly the case for executives and managers. Roughly two thirds of executives (67\%) and managers (64\%) believe generative AI can help them communicate more effectively. • Early adopters of ChatGPT in this sample hold much different views of generative AI than do non-users of ChatGPT. Early adopters hold the following distinctive views: o They are much more likely to think AI is good for society than non-users (64\% to 22\%) and believe it will make them more productive (82\% for early adopters; 26\% for non-users); however, they are also more likely to worry about the ethical implications of AI (68\% to 55\%) in the workplace and worry that their own job will be replaced by AI (41\% to 20\%). o They are much more likely to think generative AI will support them in their work. About 85\% of early adopters say that ChatGPT can help them generate ideas for work compared to about 50\% of non-users. About 73\% of early adopters say it can improve the quality of their work compared to 42\% of non-users. About 74\% of early adopters say it can help them communicate more effectively compared to 41\% of non-users. Executives and managers are slightly more likely to be enthusiastic about the benefits. • Employees in organizations with generative AI policies view these policies positively. Those who are aware of an organizational policy about generative AI generally believe it has supported more comfort in using ChatGPT for work, has improved trust, has improved efficiency, and has provided legal protections. Those who are early adopters are generally more positive about each of these benefits of organizational policy than those who are non-users of ChatGPT. • Most early adopters of generative AI in organizations without generative AI policies want more guidance about ChatGPT use. Most early adopters believe an organizational policy would make them more comfortable using ChatGPT (61\%), that it would increase trust (56\%), and that it would improve efficiency (66\%). Key Recommendations • Develop generative AI policies that support innovation and efficiency while putting into place legal safeguards for organizations and their employees. • Use a social contracts approach to develop generative AI policies.}, language = {en} } @article{FleischmannCardonMa2022, author = {Fleischmann, Carolin and Cardon, Peter W. and Ma, Haibing}, title = {Algorithmische Auswertung von Mitarbeiterdaten: zwischen Vertrauen und Kontrolle}, series = {PERSONALquarterly}, volume = {2022}, journal = {PERSONALquarterly}, number = {03}, year = {2022}, language = {de} } @article{CardonMaFleischmann2021, author = {Cardon, Peter W. and Ma, Haibing and Fleischmann, Carolin}, title = {Recorded Business Meetings and AI Algorithmic Tools: Negotiating Privacy Concerns, Psychological Safety, and Control}, series = {International Journal of Business Communication}, journal = {International Journal of Business Communication}, doi = {10.1177/23294884211037009}, year = {2021}, abstract = {Artificial intelligence (AI) algorithmic tools that analyze and evaluate recorded meeting data may provide many new opportunities for employees, teams, and organizations. Yet, these new and emerging AI tools raise a variety of issues related to privacy, psychological safety, and control. Based on in-depth interviews with 50 American, Chinese, and German employees, this research identified five key tensions related to algorithmic analysis of recorded meetings: employee control of data versus management control of data, privacy versus transparency, reduced psychological safety versus enhanced psychological safety, learning versus evaluation, and trust in AI versus trust in people. More broadly, these tensions reflect two dimensions to inform organizational policymaking and guidelines: safety versus risk and employee control versus management control. Based on a quadrant configuration of these dimensions, we propose the following approaches to managing algorithmic applications to recording meeting data: the surveillance, benevolent control, meritocratic, and social contract approaches. We suggest the social contract approach facilitates the most robust dialog about the application of algorithmic tools to recorded meeting data, potentially leading to higher employee control and sense of safety.}, language = {en} } @article{GetchellCarradiniCardonetal.2022, author = {Getchell, Kristen M. and Carradini, Stephen and Cardon, Peter W. and Fleischmann, Carolin and Ma, Haibing and Aritz, Jolanta and Stapp, James}, title = {Artificial Intelligence in Business Communication: The Changing Landscape of Research and Teaching}, series = {Business and Professional Communication Quarterly}, volume = {85}, journal = {Business and Professional Communication Quarterly}, number = {1}, doi = {10.1177/23294906221074311}, pages = {7 -- 33}, year = {2022}, abstract = {The rapid, widespread implementation of artificial intelligence technologies in workplaces has implications for business communication. In this article, the authors describe current capabilities, challenges, and concepts related to the adoption and use of artificial intelligence (AI) technologies in business communication. Understanding the abilities and inabilities of AI technologies is critical to using these technologies ethically. The authors offer a proposed research agenda for researchers in business communication concerning topics of implementation, lexicography and grammar, collaboration, design, trust, bias, managerial concerns, tool assessment, and demographics. The authors conclude with some ideas regarding how to teach about AI in the business communication classroom.}, language = {en} }