Teaching GenAI › Use cases
Practical applications of generative AI, each tied to research I have done with colleagues or to projects my students built at the University of Washington Foster School of Business. Follow the links for the full papers.
Companies run open innovation challenges to collect ideas from crowds. Generative AI can now draft many candidate solutions in minutes, which changes who generates ideas and what the crowd is for.
Reviewing hundreds of startup pitches, grant proposals, or internal ideas is slow. Generative AI can write structured assessments that help human reviewers decide, as tested in a field experiment.
Ordering and inventory decisions suffer from delays and the bullwhip effect. Large language models can act as planning agents, and people can work alongside them on each order.
Optimization models schedule, route, and allocate, but building them takes experts. LLMs can translate a business problem into a model, explain the solution, and let managers ask what-if questions.
When people ask AI for advice on hard judgment calls, the AI can sway the outcome. Knowing how much it influences people matters for any team using AI assistants in sensitive decisions.
Business students with no engineering background now ship working websites, dashboards, and AI agents in weeks. The same tools also make many products look alike.
Instead of sending learners to a general chatbot, a Socratic tutor can sit inside the quiz itself, guiding without giving answers and leaving a data trail for instructors.
Sources
Patient data comes as notes, images, time series, and tables. Multimodal AI combines them into one prediction, and LLMs can help build quality-of-care measures.