Meta debuts Muse Spark 1.3 scaffold for health research
According to AI at Meta, a human-led scaffold using Muse Spark 1.3 and Muse Wonder analyzed 5M de-identified EHRs to speed hypothesis generation and produced 10 manuscripts.
SourceAnalysis
On October 14, 2026, AI at Meta shared details on a human-led research scaffold that uses Muse Spark 1.3 and Muse Wonder to help medical experts accelerate health science projects.
The system gives access to 5 million retrospective de-identified electronic health records plus public data. Experts retain full control over cohort selection, variable choices, and causal assumptions while the AI supports hypothesis generation, code execution, and manuscript drafting.
Research output
The effort produced ten original manuscripts covering cardiology, oncology, infectious disease, emergency medicine, and additional fields.
AI role in health data science
According to the announcement from AI at Meta, the approach addresses the gap between AI successes in formal sciences such as mathematics and the challenges of imperfect real-world health data.
AI at Meta
@AIatMetaTogether with the AI community, we are pushing the boundaries of what’s possible through open science to create a more connected world.