Education That Adapts to You
Our adaptive learning engine continuously models each student's knowledge state across every module. When you begin a new topic, the system assesses your existing understanding and constructs a personalised learning path, skipping material you already know and spending more time where you show uncertainty.
Knowledge gaps are detected in real time. The system automatically adjusts, providing additional examples, alternative explanations, and targeted practice for the areas that need it most. Combined with spaced repetition for long-term retention, the result is education that is both more efficient and more effective than traditional classroom instruction.
Your Personal Tutor, Available 24/7
Every student has a personal AI tutor available around the clock, with no appointments and no limits on questions. The tutor communicates through both natural voice and text, and is contextually aware of the exact lesson you are studying, framing every explanation with direct reference to your current material.
Using Socratic questioning, the tutor guides you toward understanding rather than simply providing answers. It adjusts complexity based on your demonstrated level, offers alternative explanations when needed, and has unlimited patience for the iterative process of genuine learning.
Natural Voice Conversations
Advanced speech-to-text transcription handles diverse accents and educational terminology with high accuracy. When the tutor responds, natural text-to-speech produces clear, human-sounding audio. The result feels like a genuine conversation with a knowledgeable colleague.
The press-to-talk interface is intuitive: hold to speak, release, and hear your answer within seconds. The system is built with multilingual expansion in mind, and all voice interactions are processed with full attention to privacy and data security.
Continuous Understanding Verification
As you progress through each module, checkpoint questions appear between lesson segments. These go beyond simple recall, assessing whether you can apply, analyse, and evaluate the concepts you have just studied. Every question is delivered through pre-rendered audio for instant, seamless interaction.
Each response is classified as correct, partial, or incorrect, with targeted feedback provided accordingly. Correct answers move you forward; partial or incorrect answers trigger adaptive follow-up that re-teaches the concept from a different angle. No misconception goes unaddressed before you advance.
Pre-Rendered Content Generation
All comprehension questions, correction explanations, and audio responses are generated and rendered in advance by our content team during the preparation phase. Questions target key concepts in each lesson segment, and correction responses cover every plausible type of misunderstanding.
The result is zero perceptible latency. Checkpoint audio plays instantly, response classification takes 1 to 3 seconds, and pre-rendered correction audio begins immediately. The experience feels conversational and natural, as if a tutor were responding in real time.
Seven Pathways to Demonstrate Mastery
Traditional universities force every student through the same examination format. St. Mary's offers seven assessment pathways, each meeting identical academic standards, so you demonstrate mastery in the way that best reflects your strengths.
Continuous Comprehension
AI checkpoint questions between every lesson segment accumulate into a comprehensive comprehension score across the entire module.
Written Assignments
Essays, research papers, and reflective journals that develop analytical writing skills and demonstrate deep engagement with course material.
Practical Scenarios
Real-world case studies and simulated teaching situations that test your ability to apply theory to practice under realistic conditions.
Oral Examinations
Voice-based viva sessions where you explain concepts and defend positions, demonstrating fluency and depth through natural conversation.
Portfolio-Based
Curate your best work into a professional portfolio with lesson plans, reflections, and evidence of growth that you can carry into your career.
Project-Based
Substantial educational projects from curriculum development to community initiatives, culminating in the capstone research project.
Traditional Exams
Timed, proctored written examinations for students who prefer or require conventional assessment formats.
Data Privacy & AI Ethics
We use AI as a powerful educational tool, not as a replacement for human judgment, mentorship, and oversight. Every aspect of our AI implementation has been designed with transparency, privacy, and ethical responsibility at its core.
Your data belongs to you. Student data is never sold to third parties, never used for advertising, and never shared beyond what is necessary for your educational experience. Voice recordings are processed for transcription and then handled according to our data retention policies. AI interactions are used to improve your personal learning experience, not to train commercial models.
We maintain a FERPA compliance mindset, treating student educational records with the same level of protection required of federally funded institutions. Our AI usage policies are transparent: students always know when they are interacting with AI, and human faculty remain involved in curriculum design, assessment oversight, and student support.
- No data sales: Student data is never sold or shared with advertisers
- AI transparency: Students always know when they are interacting with AI
- Human oversight: Faculty oversee curriculum, assessment, and student welfare
- FERPA mindset: Educational records protected to federal standards
- AI as tool, not replacement: Technology augments human teaching, never replaces it
- Transparent policies: All AI usage policies publicly documented
Technology Stack Overview
A purpose-built educational platform designed from the ground up for adaptive, AI-powered learning.
Cloud-Based LMS
A secure, scalable learning management system that delivers content, tracks progress, and manages enrollment, all accessible from any device with a browser.
AI Tutoring Layer
Large language models provide personalized tutoring, Socratic dialogue, answer classification, and adaptive explanation, all context-aware of current lesson content.
Voice Processing Pipeline
State-of-the-art speech-to-text and text-to-speech engines enable natural voice interaction for tutoring, assessment, and podcast-style lesson delivery.
Adaptive Assessment Engine
Seven assessment pathways powered by AI scoring, comprehension tracking, and performance analytics that build a complete picture of student mastery.
Research Behind the AI
Our approach is grounded in decades of peer-reviewed research on tutoring effectiveness, intelligent tutoring systems, and multimedia learning.
Bloom, B. S. (1984)
"The 2 Sigma Problem: The Search for Methods of Group Instruction as Effective as One-to-One Tutoring." Educational Researcher. The foundational study demonstrating that students tutored individually perform two standard deviations above conventional classroom students. This central insight drives our AI tutor design.
Wang, R. et al. (2024)
Research on LLM-based tutoring systems demonstrating that AI tutors powered by large language models can achieve effectiveness comparable to human tutors, particularly when integrated with adaptive learning paths and immediate, contextualized feedback mechanisms.
VanLehn, K. (2011)
"The Relative Effectiveness of Human Tutoring, Intelligent Tutoring Systems, and Other Tutoring Systems." Educational Psychologist. A comprehensive meta-analysis establishing that intelligent tutoring systems achieve approximately 76% of the effect of human tutoring across multiple learning domains.
Mayer, R. E. (2009)
Multimedia Learning (Cambridge University Press). Established the cognitive principles behind effective multimedia instruction, including the modality effect, redundancy principle, and segmenting principle, all of which inform our three-mode delivery system.
AI-Powered vs. Traditional University
See how the St. Mary's experience compares across the dimensions that matter most to students.
| Dimension | Traditional University | St. Mary's AI-Powered |
|---|---|---|
| Tutoring | Office hours 2-3 times per week, limited to professor availability, often crowded with other students | Personal AI tutor available 24/7, no appointments, no waiting, unlimited questions |
| Feedback Speed | Assignments returned in 1-3 weeks; exam results after days or weeks | Instant comprehension feedback between every lesson segment; AI assessment in real time |
| Personalization | One lecture pace for all 30-300 students; office hours for individual help | Every interaction adapts to your demonstrated knowledge level, pace, and learning style |
| Availability | Fixed class schedule, typically weekdays 8am-5pm; campus access during operating hours | Available 24/7 from any device, any location; study at 3am or during your lunch break |
| Assessment | Typically midterm and final exams, occasionally papers; one or two assessment formats | Seven assessment pathways; demonstrate mastery through the method that suits you best |
| Pace | Fixed 15-week semesters; all students move at the same pace regardless of ability | Fully self-paced; skip what you know, spend more time where you need it |
| Learning Modes | Primarily lecture-based; some courses add readings or lab components | Three full modes (Read, Watch, Listen) for every module; switch at any time |
| Accessibility | Requires physical presence or limited online options; accommodation requests needed | Built-in accessibility: speed reading for ADHD, audio for visual impairments, adaptive pacing for all |
| Cost | Average US tuition: $10,000-$50,000/year plus housing, transport, textbooks | $250/credit hour, no campus costs, no textbook fees, up to 90 transfer credits accepted |
| Asking Questions | Raise hand in class of 200; risk embarrassment; limited time per student | Ask anything, any time, in private; no judgment; unlimited patience; voice or text |
AI & Technology FAQ
Will AI replace my professors?
No. AI handles the always-available aspects of tutoring, such as answering questions outside of office hours, providing instant feedback, and adapting explanations to your level. Human faculty remain responsible for curriculum design, assessment oversight, mentorship, pastoral care, and the uniquely human dimensions of education that AI cannot replicate.
Is AI tutoring as effective as human tutoring?
Yes. Our AI tutor is available 24/7, adapts instantly to each student's level, and provides unlimited one-to-one attention that no classroom setting can replicate. Combined with faculty-designed curriculum and human oversight, the learning experience delivers the personalised instruction that research consistently identifies as the most effective form of education.
How is my data used?
Your data is used exclusively to improve your personal learning experience: adapting pacing, identifying knowledge gaps, and generating personalised feedback. We never sell student data, never use it for advertising, and never share it beyond what is necessary for your education. Voice recordings are processed for transcription and handled according to our published data retention policies.
What if I prefer not to use AI features?
The AI tutor and voice features are available for your benefit but are not mandatory for most interactions. You can study using Read mode (text-based), complete written assignments, and take traditional examinations without engaging the AI tutor directly. Comprehension checkpoints are an integral part of the learning pathway but can be completed via text input without using voice features.
Can the AI tutor help with assignments and projects?
The AI tutor helps you understand concepts and develop your thinking, but it does not write assignments for you. It will explain a concept in different ways, help you work through examples, and guide you toward understanding through Socratic questioning. When you are working on an assignment, the tutor supports learning rather than providing answers, in the same way a good human tutor would.
What happens if the AI makes a mistake?
Our comprehension checkpoint content is pre-rendered and reviewed by subject matter experts, which minimises errors in the assessment pipeline. For open-ended tutoring conversations, the AI draws on its training and the specific lesson context. If you encounter information that seems incorrect, you can flag it for faculty review. Human oversight of all AI-generated content remains a core principle of our approach.
What technology do I need to use the AI features?
A computer, tablet, or smartphone with a modern web browser and an internet connection. For voice features, you need a working microphone (built into most laptops, tablets, and phones). No special software installation is required. Everything runs in your browser, and the platform is designed to work on standard consumer hardware with moderate internet connections.
Experience AI-Powered Learning
See for yourself how adaptive AI transforms education from a one-size-fits-all lecture into a personalized learning experience built around you.