The Companion experience includes two AI-powered assistant experiences: the Companion AI Assistant and the Companion Course Assistant. The Companion AI Assistant provides broad in-app support to help students understand general information, navigate Companion, and resolve common questions about processes and terminology. The Companion Course Assistant is a course-focused assistant designed to help students with course-related information, such as assignments and due dates, using student-specific academic context.

Together, these assistants are designed to make support faster, clearer, and more accessible inside Companion. Instead of relying only on static help content or waiting for human support, students can ask questions in natural language and receive contextual guidance in real time. This helps reduce confusion, improves student autonomy, and creates a more intuitive support experience across both general and course-related needs.

The Companion AI Assistant and the Companion Course Assistant are designed to provide faster, clearer, and more accessible support within Companion. Together, they help students resolve common questions more independently while improving the overall efficiency of student support.

Main Objective

Provide real-time, in-app support that helps students understand information, navigate processes, and act on both general and course-related questions with greater confidence.

Expected Outcomes

• Reduce repetitive support questions and basic mentor/support escalations.

• Improve student understanding of academic and administrative processes.

• Improve access to course-related information such as assignments and due dates.

• Increase student autonomy when navigating Companion and related systems.

• Help students resolve common questions any time, any day (not only during business hours).

• Improve the student experience by delivering faster, more accessible guidance.

• Allow mentors and support teams to spend more time on complex or high-impact cases.

• Establish a strong AI support foundation in Companion that can continue evolving as new support capabilities are developed.

The Companion AI Assistant is currently in development, and the team is preparing an MVP for internal Digital Ops testing while the required Azure permissions are still pending. The Course Assistant is also in development as a course-focused assistant experience within Companion and depends on the same backend resource readiness before broader testing can move forward.

In parallel, the team is preparing stakeholder demo materials and internal validation scenarios. Once Azure resources and permissions are available, the next step will be internal testing in the Companion app, followed by planning for broader rollout stages.

1.- What is the current status of the AI Assistant?

The Companion AI Assistant is currently in development. The team is preparing an MVP for internal Digital Ops testing while waiting for the required Azure permissions to enable a broader internal test inside the Companion app.

2.- When is the Course Assistant planned to be available for students?

There is no student release date yet. The team cannot plan a timeline until the required Azure resources and permissions are available.

3.- Are both assistants going to be accessible in Native and Web?

At this time, the Course Assistant is planned for the Companion app (Native). The Companion AI Assistant is expected to be available as an updated version of the current Companion chatbot experience, including broader Companion usage (app and web).

4.- Where does the Companion AI Assistant and Course Assistant get their information?

The Companion AI Assistant is designed to provide guidance using trusted BYU-Pathway-related sources, such as main BYU-Pathway websites and other official resources (for example, EnglishConnect, the catalog, and the Help Center). Its role is to help students understand and navigate general information across the Companion experience.

The Course Assistant is designed to use student-specific academic information from Dataverse, focused on course-related data. It does not perform open-ended searches directly in Dataverse. Instead, the AI generates the parameters needed for a controlled lookup, and the system returns only the specific information required for the response.

5.- Does the AI search directly in Dataverse?

No. The AI does not directly search Dataverse on its own. The system uses controlled filters and parameter-based queries, then returns only the relevant data needed for the assistant’s response. This helps keep the experience more structured and controlled.

6.- Will having these assistants in Companion make the app or web experience heavy?

The assistant experience is not expected to make the Companion app or web experience heavy. Most of the processing happens in backend services (such as Azure/Copilot resources), while the app mainly provides the chat interface. In practice, the Companion side remains lightweight because the heavy work is handled outside the app.

7.- Can students use the assistants without internet?

No. The assistants require an internet connection because they rely on connected AI and backend services. They are not designed for offline use.

8.- Do students need very strong internet to use the assistants?

Students do need internet access, but the Companion experience is designed to stay lightweight on the device because the app mainly handles the chat interface while the heavier processing happens in backend services. This approach helps reduce the load on the student’s phone, including older devices.

9.- Will conversations be saved in the first versions of the assistants?

The first versions are expected to use Copilot Studio, and conversation history is not planned to be saved in those initial versions.

10.- Why won’t conversation history be saved at first?

This is mainly due to how the initial Copilot Studio-based version works. It is a faster way to launch the first version, but it comes with some limitations, including how conversation history is handled.

11.- Could conversation history be added later?

Yes. Saving conversation history is possible in the future, including through BYU-Pathway-managed servers if needed. It is not a storage or performance limitation; it is mainly a product and implementation decision for the first version.

12.- What is the tradeoff of using Copilot Studio in the first version?

The main advantage is speed of delivery, since it helps the team launch faster. The tradeoff is that the team is more limited by the features and constraints available in Copilot Studio for the first version.

13.- Is the Course Assistant a separate product from the AI Assistant?

No, they serve different purposes, but both belong to the Companion experience. The Course Assistant is a course-focused assistant, while the Companion AI Assistant provides broader support across Companion. Right now, the Course Assistant is being built as a separate bot experience within Companion, and the team is still evaluating how these experiences may be combined or presented in the future.

14.- In what languages will the AI Assistant and Course Assistant be available?

The official language is English. However, because the assistants are powered by an AI large language model (LLM), they are expected to respond in the user’s language as part of the model’s multilingual capability.

15.- What will the Course Assistant NOT do?

The Course Assistant is not designed to replace mentors, instructors, or support teams, and it will not make academic judgments such as deciding which assignment is most important or what a student should work on first. It also cannot reliably evaluate assignment difficulty or estimate how long a task will take, especially when assignment instructions are stored in external resources that are not fully available to the assistant. In addition, it is not intended to create custom reminders or schedule personal study plans; its calendar action is limited to adding assignment due dates.