UT Sage
UT Sage Tutors
UT Sage is a virtual instructional designer and AI Tutor designed by UT Austin experts to support the responsible use of generative AI in teaching and learning. Instructors can use UT Sage to create interactive chatbots that enhance student engagement and customize learning experiences. UT Sage allows instructors to design tutor sessions on any topic using established principles of learning experience design, while also ensuring student education records are protected. AI tutors coach students in specific topics or lessons related to their coursework.
We are currently in an open comment period for UT Sage. If you have feedback, questions or other comments about the development of Sage, please use the open comment form.
Why Use Sage?
First and foremost, you can ensure that the privacy, security, and copyright and intellectual property rights of you and your students are protected because UT Sage meets the compliance and security guidelines of the University.
In addition, with Sage, faculty train the tutors to customize and tailor the experience for their students. When you create a new tutor, UT Sage will ask a series of questions that will train a custom chatbot using established principles of effective learning experience design, such as starting out with defining what you want students to know and be able to do after engaging with UT Sage. In addition to answering questions about your students, learning outcomes, common misconceptions and activities, you can upload resources that UT Sage will use as the first and best source of information about a topic.
Getting Involved with UT Sage
We are currently in the early pilot and development phase of Sage. This means we are testing prototypes of Sage with a small set of instructors and students and making improvements. In the spring, we will increase the size of the pilot. If you are interested in piloting and evaluating Sage, sign up for more information.
Frequently Asked Questions
UT Sage is currently in beta testing with a small number of faculty and students. If you are interested in piloting and evaluating Sage, sign up for more information.
When you create a new tutor, UT Sage will ask a series of questions that will train a custom chatbot using established principles of effective teaching. In addition to answering questions about your students, learning outcomes, common misconceptions and activities, you can upload resources that UT Sage will use as the first and best source of information about a topic.
This is a possible outcome for all AI models, not just UT Sage. Part of responsible engagement with generative AI means being aware of its limitations. Transparency is key. A clear disclaimer about the limitations of generative AI is always present within UT Sage’s interface to remind users of this fact. Just like human tutors, generative AI tools can sometimes miss the mark with construing information. Evaluating information is an important AI literacy skill, and we recommend all instructors teach students the limitations related to accuracy before an AI tutor is introduced as a course resource. Review the Office of Academic Technology’s Addressing the Limitations of Using Generative AI for Learning Guide for more information.
Yes, you can upload your syllabus to Sage, and it will interpret the information for your students.
UT Sage is a platform that is approved for use only by The University of Texas at Austin, which means that your (or your students’) data will be kept securely on the platform. The University does not share any of the resources or interactions with UT Sage with third parties to train LLMs or for marketing purposes.
UT Sage uses the large language model (LLM) Llama 3.1 405B to understand what you say and how to respond. It also uses retrieval-augmented generation (RAG) to learn more about your topic and course.
The platform is a collaboration between the Office of Academic Technology and Instructional Technology Services with backend development from AWS.
