88%
Highest endocrinology benchmark score
Our Health Context Engine outperforms GPT, Gemini, and human clinicians on the endocrine medical board exam. Its proprietary RAG reaches higher accuracy without web access to guidelines and literature.
Built On Science. Powered By AI.
January has been building glucose prediction models since 2020 and publishing the results. This page is the evidence file: benchmarks, methodology, the research record, and the people who hold us to it.
88%Endocrinology board exam accuracy
0.8%Hallucination rate on clinical cases
Patented techPatent portfolio
Measured against the exam clinicians actually sit
0.8%
Hallucination rate
Against 1.8% contradictions, materially below general-purpose frontier models on the same clinical cases (−log₁₀(p) = 8.0 and 8.4).
33h
Forward glucose prediction
CGM-based glucose predictions 33 hours into the future, and virtual continuous glucose curves for the 128M people who will never wear a sensor.
Behavior change that reaches the threshold that matters
Protein Intake
+6.9%
Carb Intake
-3.5%
Fiber Intake
+13.8%
Sugar Intake
-16.7%
Participants lost weight
+74.4%
Time in Healthy Glucose Range for people with T2D *
49.7% => 57.4%
* Every 5% increase in Time in Range is clinically meaningful for people with type 2 diabetes.
Six years of models, in order
2020
CGM-based glucose predictions 33 hours into the future
2021
Accurate post-CGM predictions
2022
Virtual continuous glucose curves
2023
Millions of data points from thousands of patients inform a probabilistic model without sensors
2025
AI Health Context Engine built on a medical-grade personal health knowledge graph
2026+
Integrated patient application and clinical decision support tools
Peer-reviewed papers
2025 npj Digital Medicine
AI Health Context Engine on a medical-grade personal health knowledge graph
Read the paper 2023 npj Digital Medicine
Glucose prediction and behavior change without continuous sensors
Read the paper 2021 Diabetes Therapy
Digital program outcomes for people living with type 2 diabetes
Read the paper 2025 ADA Scientific Sessions
Evaluating a flexible, AI-supported digital diabetes program
Read the paper 2021 IEEE
Machine-learning blood glucose prediction from CGM data
Read the paperMethodology reviewed by our AI and Scientific Advisory Boards. Meet them