Works across a range of course types and teaching styles.

Support for learning beyond scheduled class time — helping students prepare before lessons and reflect afterward.

• ·Individual learning. Students explore concepts at their own pace, test their understanding, and get guided prompts aligned with the course material.
• ·Self-guided collaboration. Unsupervised small groups structure discussions, compare ideas, and resolve uncertainties together.
• ·Group work. Shared inquiry that makes reasoning visible, so groups build on each other's thinking rather than dividing tasks.

In live classroom settings, ChatTutor becomes part of the instructional flow.

• ·Supervised group work. Educators guide how AI is used while students work through problems or case studies.
• ·Small classrooms and large lectures. Deeper dialogue and personal scaffolding in small rooms; many students engaging at once through structured prompts in large ones.
• ·Assignments and exams. In development Learning-focused assessment that emphasises reasoning, transparency and academic integrity.

Built for human + AI collaborative learning, where teachers, students and AI share the same learning space.

• ·Agent participation. Educators decide when AI takes part in a discussion thread and when it stays in the background as a support tool.
• ·Peer validation. Peers or teaching assistants can be looped into a conversation to check reasoning and correct misunderstandings.
• ·Curriculum control. Educators decide how AI is applied to their curriculum, so it reinforces pedagogical goals rather than dictating them.

Meaningful insight into how students are actually learning.

• ·Weekly bottlenecks. Where students struggle, or ask the same question again, surfaces week by week.
• ·Depth, not activity. Performance-level indicators separate surface engagement from deeper conceptual mastery.
• ·Mid-semester correction. Instructional adjustments and targeted support while the semester is still running.

For teams who want to go further, agents can be built from scratch.

• ·Custom agents. Built for a specific subject or teaching strategy.
• ·AI personas. A Socratic tutor, a debate partner, a writing coach.
• ·Reasoning first. Agents that probe assumptions and ask students to explain their thinking instead of producing answers.

Reading and dialogue are only two ways into a subject. The next modalities extend the same grounded course material into other forms of study.

• ·Podcasts. Course material turned into something a student can listen to — orchestrated by the teacher.
• ·Collaborative notes. A shared record a cohort builds together, with the agent in the margin.
• ·Multiple file formats. Documents, video and more, indexed as course material the agent can answer from.

Several agents, each with a distinct way of explaining, become material for discussion in class.

• ·One agent per response style — short, detailed, step-by-step
• ·Students compare explanations of the same question
• ·Selected answers reused as classroom discussion material

Answers stay anchored in the uploaded course material, and question patterns feed back into the next lecture.

• ·Notes and articles uploaded, so answers match what is taught
• ·Analytics surface recurring misunderstandings across the group
• ·Lectures and exercises adjusted to the gaps that appear

ChatTutor plays the applicant company; students interrogate it from assigned professional roles.

• ·Full application dossier loaded; the agent answers as the company
• ·Students work as regulator, bioethicist, reimbursement specialist
• ·Findings compared in class: which questions exposed most

A shared session lets the whole room answer the same question at once — then @ai clusters the answers live.

• ·Every student answers, not just the few who speak up
• ·Typing @ai groups answers and surfaces misconceptions
• ·Contributions kept, so participation can be reviewed after class