Skip to content
Research

Insilico Medicine Open-Sources Longevity Grok, a Health AI That Scores Its Own Predictions

The multi-agent system, built with Harvard's David Sinclair, proposes health interventions and later checks whether users' biomarkers moved as forecast.

By Nakoda News Desk

·2 min read

Prefer Nakoda News on Google

Insilico Medicine has released the code for Longevity Grok, a multi-agent AI system that gathers a person's scattered health records, suggests lifestyle or nutritional changes and then grades how accurate its own forecasts turned out to be. The Hong Kong-listed drug developer announced the release on November 6.

The system, which the company also calls Personal Longevity Superintelligence, or PLSI, runs on the Grok Bot platform and was developed with David Sinclair of Harvard Medical School, Insilico said in a press release. The client software and a research backend called the Longevity Mothership are published on GitHub. The release does not name a specific open-source licence.

The scoring feature is what sets it apart from most consumer health chatbots. Each proposed intervention comes with a "prediction card" listing the biomarker it targets, a confidence interval, a time horizon and an evidence grade. After follow-up tests, the system records whether results landed inside the predicted range and tracks how well calibrated its predictions are over time.

Insilico said PLSI pulls in lab panels, continuous glucose monitor readings, wearable data, epigenetic tests and clinical files, standardising them with medical coding systems including LOINC and RxNorm. It coordinates up to 30 specialised agents, with some assigned to challenge the recommendations of others.

The company describes three uses: a personal longevity assistant, a physician's assistant that provides structured patient history while leaving diagnoses and prescriptions to licensed doctors, and a research tool. Users can opt in to share consented, de-identified data with the Mothership, and can donate spare daily AI usage to background research in the style of volunteer computing projects. Raw medical files stay on the user's side unless they opt in, Insilico said.

The company was candid about the state of the field. "Currently, there are zero tools that are clinically proven to give a healthy person an additional year of life," said Alex Zhavoronkov, Insilico's founder and chief executive. The documentation limits Longevity Grok to research and education, and the release cites no independent validation of its predictions.

The launch follows Insilico's September release of LongevityBench, a set of compact longevity language models and an agentic research platform. The company listed on the Hong Kong Main Board in December 2025.

Sources

Produced with AI assistance from the sources above. How we work

By

Nakoda News Desk

Nakoda News covers AI and business. Every story links the sources it draws on; how we work is set out in our editorial standards.

How we work

Related coverage