Ora AI · Research
The Most Evidence-Based Medical Study Platform.
Built on 1.09 billion student-learning data points and validated by the largest MedEd RCT to date.
Flagship trial
Multi-institutional RCT · 10-month real-exam follow-up
Ora AI Outperforms UWorld in the First Multi-Institutional RCT of a USMLE Question Bank.
Preregistered, blinded-analysis, multi-institutional trial of 155 medical students. At-risk students improved 2.4× more with Ora than UWorld over the 14-day intervention. At 10-month real-exam follow-up: 100% Step 1 first-attempt pass rate in the Ora arm, and Ora-exposed students scored 11 points higher on first-attempt Step 2 CK.
Read the full trial brief →USMLE Step 1 first-attempt pass
10-mo follow-up
USMLE Step 2 CK first-attempt mean
10-mo follow-up
QBank
4 briefsMastery trajectories
Medical Students Reach Sustained Topic Mastery in a Median of 10 Encounters on Ora AI's World-First Spaced-Repetition QBank, Validated Across 1.46M Per-Student Topic Records.
+6.9 pp
Accuracy gain, encounter 1→10
Spaced repetition
Students Who Fail a Vignette on Ora AI Recover the Concept 80% of the Time Within Three Spaced Variants.
80%
Failed concepts mastered within 3 spaced variants
Item-bank psychometrics
7,399 Vignette Distractors From Ora AI Exceed All Published Item-Analysis Benchmarks.
82.3%
Distractors meeting Haladyna ≥5% threshold
Cross-school benchmarks
Cross-School QBank Benchmarks Show How Each School’s Students Compare to 19 Others on Identical Content, System by System.
20 schools
Benchmarked on identical content
AI
4 briefsAI safety & grounding
Ora's AI Tutor Answers Medical Student Questions More Accurately than Frontier Models GPT 5.5 (OpenAI), Claude Opus 4.8 (Anthropic), and Gemini 3.1 Pro (Google).
4.5 / 5
Contextual relevance (vs 3.5 ungrounded)
Curriculum mapping
Faculty and Medical Students Save 15 Hours per Week Using Ora's Curriculum-Aligned AI Across 11,752 Lectures from 161 Schools.
~15 hr
Reclaimed per week per student
AI chat anchoring
Medical Students Anchor 73% of Their AI Questions to Specific Study Materials, Validated Across 11,847 Conversations.
72.5%
Of AI questions anchored to study content
Content architecture
Ora AI Scaffolds Medical Students using 3.2 Million Context-Dependent Hyper-Specific Micro-Explanation Atoms.
3.24M
Context-specific explanation atoms
Flashcards
2 briefsVideos
1 briefGames
1 briefFlashcard content currency
Ora's Physician-Reviewed AI Flashcards Have 449 Fewer Medical Errors than the Student-Crowdsourced AnKing Deck.
3.07×
AnKing vs Ora strict-error ratio
FSRS validation
Ora AI's Proprietary FSRS Algorithm Variant Beats Anki's FSRS across 969,000 Medical Flashcard Reviews, with 98.9% 1-Year Retention.
94.4%
Empirical retention (vs 90% FSRS target)
Videos made interactive
Ora AI's Active-Recall Videos Lift QBank Accuracy Up to 4.4 Points on 80,000 Same-Topic Responses.
+2.66 pts
QBank accuracy lift on same-topic items
Game-based learning
Game-Based Learning Shows Consistent Effects in Medical Education; Ora AI's Game Engagement Reflects the Same Pattern.
5
Meta-analyses + systematic reviews synthesized