W. P. Carey faculty discuss artificial intelligence in business, education, and beyond

Last month's Coffee, Tea, and ChatGPT session focused on the consequences of ungoverned technologies and the implications for teaching and learning.

The W. P. Carey School of Business introduced the speaker series "Coffee, Tea, and ChatGPT" during the spring and summer of 2023 to share the impact of generative AI and ChatGPT on teaching and learning and to bring faculty and staff together to learn from each other. The series rotates around the business school academic departments as conversation starters and involves faculty and staff across W. P. Carey, along with colleagues from other ASU units and the Office of the University Provost. This story is the first in a series about Coffee, Tea, and ChatGPT to share learnings with the broader community.

Machine learning and artificial intelligence (AI) improve efficiency, accuracy, and effectiveness in the workplace, but as these applications continue to expand in scope, it is essential to exercise caution and consider whether these emerging technologies may give rise to unintended negative consequences. Pei-yu (Sharon) Chen, Information Systems chair and Red Avenue Foundation Professor, presented on the need for mindful approaches to technology to mitigate its over-reliance and misuse at W. P. Carey's ninth Coffee, Tea, and ChatGPT session for ASU staff and faculty last month.

As machine learning algorithms like ChatGPT become more intelligent and increase their learning capacity, they absorb both accurate data, and biased, incorrect data — leading to the spread of disinformation or amplification of biases that negatively impact equity and inclusivity.

"You might have improved your accuracy and efficiency, but at the expense of some societal groups," said Chen, who used an example of a company that implemented AI algorithms to assist with their hiring process. The company improved the process's efficiency, but later learned that women applicants were significantly less likely to be hired due to the technology's previously undetected bias. Chen emphasized that blindly implementing AI technologies threatens vulnerable groups due to a lack of available historical data or the embedded biases, which knowingly or unknowingly, are present in the data. This influences AI's learning.

As director of W. P. Carey's Center for AI and Data Analytics (AIDA) for Business and Society (aka the Mindful AI center), Chen advocates for implementing mindful AI practices. Mindful AI is a holistic attempt to understand technology's broader impact on different groups of people, society, and humanity with the intent to reduce the risk of technology misuse, which affects ethics, fairness, privacy, security, and the well-being of people through intended and unintended consequences. The AIDA Center supports sustainable AI use through research, education, and innovative solutions that bring positive and lasting transformations in how individuals, organizations, and society view, develop, and use AI.

To effectively implement mindful AI, Chen recommends gauging an AI application's intended and unintended effects by defining the AI's goals; understanding how the AI is used, designed, applied, and governed to limit unintended privacy, security, and ethical consequences; and by recognizing the broader, long-term impacts of AI on its users, organization, society, humanity, and the planet. In her presentation, Chen also demonstrated how Mindful AI connects deeply with the newest ASU design aspiration, Practice Principled Innovation. Attendees at the November Coffee, Tea, and ChatGPT session received a card deck explaining the different elements of principled innovation from Associate Dean for Teaching and Learning Dan Gruber, and Jared Byrne, director for the Center for Entrepreneurship, who are the W. P. Carey College Catalysts tasked with bringing the design aspiration to the college of business.

"Know the process, because there are a lot of risks involved," said Chen. "You need to know the ins and outs, all the risks and biases that could potentially be there."

Following Chen's presentation, Clinical Associate Professor of Information Systems Matt Sopha shared how his colleague Kumar Sirugudi, assistant teaching professor of Information Systems, polled students to understand their perspective of generative AI and create syllabus criteria for AI use. Sirugudi's poll showed students saw the benefits of AI, but were concerned about the technology's potential misuse and the need for regulation.

"Students called out some of the same things we have been discussing collectively as a group," said Sopha.

Sopha also shared examples of his ChatGPT research, prompted by a Department of Information System's meeting about how faculty can use generative AI to improve the student experience and course preparation. Sopha demonstrated how he's used ChatGPT, DALL-E, and Adobe Firefly to generate content.

The session concluded with a question and answer session from the audience, with several faculty expressing concerns about ethical dilemmas surrounding AI use.

Exploring machine learning

Last year, Gruber partnered with W. P. Carey teaching leads to bring Coffee, Tea, and ChatGPT sessions to faculty and staff from across the university. Gruber and the teaching leads were also keen to ensure alignment with the ASU provost's office and Gemma Garcia, executive director of learning technology in the office of the provost, who has been a regular attendee of Coffee, Tea, and ChatGPT. Each session features a presentation or panel discussion by faculty working with machine learning in different capacities, and explores the challenges and benefits of applying ChatGPT and other machine-learning technologies in business, education, and the arts.

The 2023 Coffee, Tea, and ChatGPT season kicked off this summer with presentations from W. P. Carey's supply chain management, information systems, marketing, and management and entrepreneurship departments, respectively, as conversation starters. The goal was to bring discussions that were occuring in individual academic units to the entire school. Faculty exchanged ideas and solutions to concerns about technology's disruption of traditional teaching styles and affect on academic integrity issues, classroom criteria for AI use, and how ChatGPT's evolving capabilities impact student learning.

Inara Scott, senior associate dean at Oregon State University's college of business, discussed what faculty should be teaching in the classroom and how to assess student learning during the season's third session, "Rising to Meet the Challenge of ChatGPT."

"I really believe that this is a tipping point for higher education," said Scott. While some educators feel generative AI should be banned completely in academia, others anticipate it being fully embraced and integrated into all aspects of higher education.

"This is really, I think, going to change our world, it's going to change how we teach, it's going to change the way we do work," she said.

During his presentation in September, Andrew Maynard, professor at the School for Future of Innovation in Society, described spending the past two decades studying advanced technology transitions, or the use of transformative technologies — from synthetic biology, to autonomous vehicles, to AI — to positively shape society's future. Maynard, who is both a thought leader in the field and one of ASU's leading faculty members in the area, discussed the difficulties that accompany understanding quickly changing AI, saying that the challenges and opportunities society faces are about "how we begin to understand, in very novel ways, what is happening, what the technologies are that are developing, what the transformative power of them is, and how we move through those transformative natures of the technology to build the world we want to build."

At the eighth Coffee, Tea, and ChatGPT session in October, Punya Mishra, associate dean of scholarship and innovation at the Mary Lou Fulton Teachers College and affiliate faculty at The Design School, discussed a range of his creative experiments with generative AI, which featured the kinds of superpowers these new technologies provide educators and researchers. He discussed how AI's ability to make content like art, coding software, and voice recordings affects people's understanding of ownership and identity, and suggested that instead of viewing large language models as search or "truth" engines", it may be more productive to view them as an "expert psychological other" that can be prone to error or bias. Mishra is also a global thought leader in the field of education, and one of the most impactful scholars at ASU in this area.

Reflecting on the evolution of the series over the last eight months and looking ahead to 2024, Gruber noted, "I am inspired by how this idea has come to life in such a meaningful manner that is bringing people together to learn and share in ways that are helping to move us forward. The participation of experts across W. P. Carey and ASU has been energizing for all of us."

The next Coffee, Tea, and ChatGPT presentation will be in January 2024, and will be an opportunity for faculty to share what they learned in the fall semester about ChatGPT and AI, and its implications for teaching and learning, with a follow-up article sharing the lessons.