Join us for the third CFE Research Seminar for 2024 presented by Jann-Hannes Pollard, a Master’s student at the Technical University of Munich.
__________________
Date: Wednesday 17th April 2024 | 12:00-1:00pm including a light lunch
Venue: Z730, Level 7, Z Block Gardens Point
Register: future.enterprise@qut.edu.au
__________________
Seminar abstract
Generative AI received enormous attention after the release of ChatGPT at the end of 2022 (OpenAI, 2022). Working professionals quickly started using this new technology to support their work, often without informing their organizations (Glassdoor, 2023). This also led to the improper use of those tools, as some employees uploaded sensitive information to the chatbot. As many providers of generative AI tools train their models on user input, this might lead to prompt leakage, the leakage of prompt information to other users (Gorcenski, 2023). In one case, a Samsung employee uploading sensitive code to ChatGPT even led the company to temporarily ban all employees from using this technology until they could ensure safe usage (Gurman, 2023). Providers of generative AI tools such as OpenAI have already started to mitigate those risks and offer their enterprise customers generative AI tools that do not use prompt data to train on (OpenAI, 2023). There are, however, still a variety of security risks (Gorcenski, 2023), legal challenges (Gonzalez Torres, Kajava, & Sawhney, 2023), and technological limitations of generative AI (Burtsev, Reeves, & Job, 2024) organizations need to manage when adopting this technology. While there are existing approaches to manage the risks of traditional AI (Schneider, Abraham, Meske, & Vom Brocke, 2023), generative AI introduces new challenges, which makes it necessary to adjust those governance mechanisms (Minkkinen & Mäntymäki, 2023).
In this talk, the first iteration of a framework for the Enterprise Governance of Generative AI will be presented, highlighting the processes, structures, and mechanisms organizations need to have in place to maximize the enterprise value of generative AI while minimizing the risks of this technology. This framework was developed based on an extensive literature review on AI Governance and adjusting those to govern the novel characteristics and use cases of Generative AI. Following a design science approach, this framework will be refined and validated in qualitative interviews with industry practitioners.
About Jann-Hannes Pollard
Jann-Hannes Polland is a master’s student at the Technical University of Munich. He is currently a Visiting Research Student at QUT, as part of an ongoing collaboration between the KrcmarLab at the Technical University of Munich and the School of Information Systems and Centre for Future Enterprise. During his stay at QUT, he is writing his final thesis on the Enterprise Governance of Generative AI.