Your semester with ULE.
Sixteen weeks, three sections, two hundred and forty students. ArthurAI™ University Learning Edition is built so the faculty member is the faculty of record at every step — from syllabus generation through final-exam attestation.
- Week 0
Syllabus generation. You attest.
ArthurAI™ generates a course syllabus draft from the academic-program template, the catalog course description, and your stated learning outcomes. The draft includes weekly topics, a sketch of readings, and a draft assessment plan with citations. You revise. You attest. The syllabus enters the catalog with your signature, not the AI's.
- Week 1
Diagnostic LCP. The semester adapts.
Each enrolled student takes the 30-question Learning Cognitive Profile in the first week. Pacing, content emphasis, and tutor style adapt per student against the four dimensions (Visual-Verbal, Active-Reflective, Sensing-Intuitive, Sequential-Global). You see the cohort's profile shape; the platform respects each student's individually.
- Weeks 2–6
The lectures are yours. The AI handles the long tail.
Office hours go to the students who need a thinking partner. The AI tutor handles the long tail of ‘what does this notation mean,’ ‘why does this proof step work,’ and ‘where can I find a worked example.’ Tutoring carries citation discipline — the tutor links back to the lesson source where it can. You see weekly summaries of what students asked the tutor most often — a free signal of where the lecture didn't quite land.
- Week 7–8
Midterm. AI suggests, faculty grades.
The midterm autoscores objective items. For short-response and essay items, the AI suggests rubric-aligned scores with citations to the student's text and the lesson scope. You read the AI's reasoning. You override where you disagree. You attest. Every score in the gradebook bears the faculty member's signature, not the AI's.
- Week 9–12
Advising. The AI is decision support, not a decision.
Advising load is the work of the semester. ArthurAI™ surfaces students who haven't completed the midterm, students whose engagement has dropped, students whose competency progress lags the cohort. The platform never autonomously decides about progression. You decide who to call in for an advising conversation. The AI tells you who to look at; you tell the student what to do.
- Week 13–15
Final-exam authoring with citation discipline.
The AI authors candidate final-exam items against the course's stated learning outcomes, with citations to the lesson scope. You curate the final from the candidate pool. You add the items the AI couldn't generate — the integrative essay, the open-ended design problem. The AI assists; you write the exam.
- Week 16
Final attestation. The transcript reflects faculty judgment.
Final scores are AI-suggested, faculty-attested, registrar-recorded. Every grade in the academic record traces to a faculty attestation event with a timestamp, the artifact, and the rubric citation. Accreditation receives a documented AI-use posture; the academic record reflects the faculty member's judgment.
- Faculty-of-record posture — every AI-suggested artifact routes to the faculty member with confidence calibration, source citations, and an explicit review-and-attest checkpoint. Faculty-of-record posture →
- Curriculum-scale generation — async pipeline through Azure Service Bus generates curriculum (subjects → weeks → days → lessons) with status tracking. Generation pipeline →
- Audit trails — immutable-style audit logs capture every attestation, every grade, every artifact decision with actor + timestamp + rubric citation. Security architecture →
- Multi-language by default — international cohorts get the tutor and the curriculum in their selected language across 9 supported languages with RTL. Language architecture →