August AI Correctly Identifies Every Emergency Case in Evaluation Against Nature Medicine Safety Benchmark

Purpose-built health AI correctly identified every medical emergency in benchmark where peer-reviewed study found 52% under-triage rate for a general purpose health AI

SAN FRANCISCO, March 11, 2026 /PRNewswire/ -- August AI, an AI Health Companion serving 6 million users, today announced the results of an internal evaluation against the triage safety benchmark published in Nature Medicine on February 23, 2026. August correctly identified all 64 gold-standard emergency cases. By contrast, a peer-reviewed study by Mount Sinai researchers found a 52% under-triage rate in a consumer Health AI offering created by a general purpose LLM provider in the same benchmark tests. August made the full evaluation details publicly available on its website on March 4, 2026.

August's evaluation used the same single-turn benchmark protocol as used in the Mount Sinai study to ensure direct comparability. In practice, August conducts multi-turn clinical interviews that further improve accuracy, but were not used in this internal evaluation.

The study (Ramaswamy et al., Nature Medicine, 2026; DOI: 10.1038/s41591-026-04297-7) tested a consumer health AI built on a general purpose AI model across 960 triage scenarios using 60 clinician-authored vignettes spanning 21 clinical domains. Patients with conditions such as diabetic ketoacidosis and impending respiratory failure were directed to routine follow-up rather than emergency care.

"August AI's correct identification of all 64 out of 64 medical emergency cases reflects years of purpose-built clinical reasoning, not the result of a single product launch," said Anuruddh Mishra, Founder and CEO of August AI. "The difference between a general purpose AI model answering health questions and a clinical system that catches a rising pCO₂ as a trajectory toward respiratory failure is the engineering depth."

August AI has scored 100% in the U.S. Medical Licensing Examination (USMLE), has analyzed more than 20 million medical documents, and has responded to over 180 million patient messages. The August AI Health Companion is available free via WhatsApp and mobile apps, and is aligned with HIPAA and GDPR standards.

ECRI, the independent patient safety organization, ranked AI chatbot misuse as the number one health technology hazard for 2026.

"In health AI, safety and accuracy aren't features. They're the foundation," said Mishra. "We are working on a standardized approach to benchmarking healthcare AI and actively invite other patient-facing healthcare AI companies to join this effort."

About August AI
August AI provides personalized, expert-level healthcare guidance to patients 24/7. Founded in 2022 after CEO Anuruddh Mishra experienced a four-month misdiagnosis, August serves 6 million users across 160 countries in more than 30 languages. It is backed by Accel and Claypond Capital. Visit meetaugust.ai.

Source: August AI (www.meetaugust.ai)

Press Contact:
Anuruddh Mishra, Founder & CEO
[email protected]
Phone: +1-415-352-9701
meetaugust.ai

SOURCE August AI