Google said on September 30, 2026 that a flu forecasting model built with its AI was the best at forecasting flu-related hospital admissions during the 2025-26 season, citing an announcement this week by the Centers for Disease Control.
What Google said
In a post by Zahra Shamsi of Google Research, the company said the CDC's end-of-season analysis published this week showed that, of 39 eligible models, Google's best matched the season's observed hospital admissions.
Google said the result "validates our confidence that AI combined with human ingenuity will improve our ability to forecast diseases worldwide."
How FluSight works
According to Google, the CDC's FluSight runs from October through May and combines weekly submissions from government, industry and academic teams predicting U.S. hospital admissions for the current week and three weeks into the future.
Google said the CDC uses the combined forecast each week to communicate the anticipated state-level demand for medical services.
The AI tool behind the forecasts
Google said its forecasts were developed using Empirical Research Assistance, or ERA, which it described as an AI tool that generates optimization algorithms across a range of scientific fields.
Research on ERA was recently published in the journal Nature, according to Google. The company said ERA's underlying technology is now available to trusted testers as part of its experimental science tools.
Key facts and where they come from
- Google says its AI-built model was the best at forecasting flu-related hospital admissions in the 2025-26 season, per a CDC announcement.
A flu forecasting model built with Google AI was the best at forecasting flu-related hospital admissions during the 2025-26 season, the Centers for Disease Control announced this week.
- The CDC's end-of-season analysis covered 39 eligible models.
The end-of-season analysis published this week shows that of 39 eligible models, Google's best matched the season's observed hospital admissions.
- FluSight runs October through May and forecasts the current week plus three weeks ahead.
From October through May, the CDC's FluSight combines weekly submissions from government, industry, and academic teams predicting U.S. hospital admissions for the current week and three weeks into the future.
- The forecasts were produced with Empirical Research Assistance, an AI tool for generating optimization algorithms.
Google forecasts were developed using Empirical Research Assistance, an AI tool that generates optimization algorithms across a range of scientific fields.
- ERA research was published in Nature and its underlying technology is open to trusted testers.
Research on ERA was recently published in the journal "Nature." ERA's underlying technology is now available to trusted testers as part of our experimental science tools.