REAL AI in Learning & Teaching RFI Response.
Campus AI's response to QUT RFI #00013003.0 — addressing all four capability lots with an enterprise SaaS platform purpose-built for Australian higher education.
Coverage
Campus AI addresses all four capability lots identified in QUT's RFI. Each lot represents a core pillar of our platform architecture, with production-ready capability available today. Canvas LTI 1.3 integration is planned and in development — see the roadmap below.
Lot 1 — Immersive Simulation & Role-play
LiveConversational and multimodal AI agents for skills practice in safe, repeatable environments. Supports clinical, legal, communication, journalism, and oral assessment scenarios with configurable personas, difficulty levels, and rubric-aligned feedback.
Lot 2 — Personalised Learning & Study Support
LiveAt-scale AI assistants grounded in course content. Adaptive support for concept clarification, course navigation, and on-demand explanation across disciplines, with academic integrity guardrails and educator controls.
Lot 3 — Formative Feedback & Writing Support
LiveRubric-aligned feedback on student drafts with pre-submission writing support. Designed for formative, non-summative use with full audit logging, AI transparency, and human-in-the-loop review.
Lot 4 — Learning Design & Content Creation
LiveStaff-facing tools for designing learning activities, assessment items, AI agent orchestration, and multi-format media content. Generated content is owned by the institution.
Data Sovereignty
100% Australian data residency. All data at rest in AWS Sydney (ap-southeast-2). AI inference in Google australia-southeast1. Contractual no-model-training clause.
Canvas LTI 1.3 (Planned)
Canvas LTI 1.3 integration with Azure AD SSO is on our roadmap (Q4 2026–Q3 2027). Planned features include course context passing, gradebook pass-back, assignment sync, and quiz integration. Today, the platform is accessible as a standalone service with full content upload and management capabilities.
Academic Integrity
Formative-only design. Full audit logging (7-year retention), human-in-the-loop review, AI content flagging, and educator visibility into all student AI interactions.
Key Design Decisions
Formative, Non-Summative Architecture
Campus AI cannot produce submittable assessment work. Feedback is advisory, flagged as AI-generated, and requires staff review before reaching students. This design point directly addresses QUT's academic integrity requirements.
Institution-Owned IP
All content uploaded by institutions and all AI-generated outputs are owned by the institution. Campus AI does not claim IP over generated content. No customer data is used for model training. Contractual safeguards in place.
Staff Self-Service
Academic staff can independently create AI agents, upload course resources, configure prompts, define rubrics, and manage course-specific AI experiences without supplier involvement — addressing QUT's self-service expectations.
Curriculum Grounding
AI responses can be constrained to QUT-approved materials only, or can leverage general foundation model knowledge where appropriate. Response sources can be attributed and distinguished — addressing Clarification 3's knowledge boundaries question.
Explore the platform.
Review our security architecture, Canvas integration details, and institutional onboarding guide.