More than 20,000 Arizona State University employees have access to CreateAI, which the university launched across ASU in 2025, and they have created over 7,000 experiences inside the platform, from simulated patients for medical training to conversational language tutors. ASU faculty, researchers and staff can use CreateAI Builder at no additional cost, starting from scratch or choosing from ready-made templates. Students can access and interact with AI experiences through faculty or staff sponsorship, which lets them work with existing AI projects and develop their own AI applications.
CreateAI gives access to more than 50 large language models, including AWS Nova, OpenAI GPTs, Google Gemini and Meta Llama. CreateAI Compare helps users choose among them, and a green-leaf icon identifies more energy-efficient models. The CreateAI Platform enables ASU's engineers and developers to build enterprise-level solutions. All data remains within ASU's governed environment and is FERPA-compliant; it is not used to train external AI models. The article says developing the in-house tool helps the university meet the demands of AI at scale while addressing cost, privacy and capability, alongside partnerships with Amazon Web Services and OpenAI.
One of the most widely adopted templates is Syllabot, a chatbot that helps students find answers to common course questions such as due dates, assignments and grading policies. There are nearly 1,000 active Syllabots at ASU, and this semester more than 300 W. P. Carey School of Business faculty members were trained to build their own. "The reality is that AI must be embedded in everything we do," says Ohad Kadan, Charles J. Robel Dean and W. P. Carey Distinguished Chair in Business.
At the John Shufeldt School of Medicine and Medical Engineering, a team is developing AI avatars that simulate patient-provider interactions. An agentic workflow called Case Creator uses multiple agents working in sync to create a patient's medical background, current condition and previous clinical appointments. The team then uses multimodal AI systems across video and audio to build avatars that simulate standard patient encounters, including emotional responses, and AI generates an evaluation that scores an interaction across 25 unique areas.
The AI Acceleration team within ASU Enterprise Technology is developing building blocks for a personalized AI experience with a single interface where a student could ask about an assignment, check on financial aid, find campus hours and email a professor without switching platforms. On the longer horizon, the article describes a personalized AI companion capable of autonomous task completion.
Details