Centaur Announces $15M Series A

‍

We are humbled and excited to share our recent $15M Series A funding round led by Matrix Partners! We’re thankful to Matrix and our other investors in this round including Accel, Global Founders Capital, Susa Ventures, Y Combinator, Omega Venture Partners, and other individual investors. We’d also like to thank all of our advisors, partners, and customers for their support.

‍

Our beginnings

‍

One of our core values at Centaur Labs is “Every Voice Counts.” This value is not only central to our company culture but also to our company history, our DNA. Our approach is based on Erik's PhD research at MIT’s Center for Collective Intelligence. The idea is that multiple opinions combined intelligently are going to be more accurate than any single opinion alone. Applying this theory to medicine and AI, we’re leveraging a network of medical experts and performance assessments to label training data. Our mission is to annotate the world’s medical data accurately so that medical AI can make the impact it’s destined to. When we first started about 2 years ago, we were collecting ~160,000 opinions per week. Today, we’re at 1 million!

‍

Our goals moving forward

‍

Our Series A funding enables us to continue our work by growing our team (engineering, data science, marketing, and sales) and continuing to invest in our network of medical experts.

‍

We’re thrilled to continue on our journey to build a global network of people and machines that are trusted based on their performance to solve medical problems.

‍

December 3, 2020

Open Source Datasets for Medical AI Training | Centaur AI

Access dozens of open-source medical AI image datasets in formats like X-ray, CT, MRI, Ultrasound, Whole Slide Imaging, and more for research and training.

November 24, 2025

High-Quality Annotation Powers Radiology AI | Centaur AI

Radiology AI improves acquisition, processing, interpretation, reporting, and long-term monitoring, but performance depends entirely on high-quality annotations. Centaur.ai delivers expert-reviewed, rigorously validated radiology labels at scale, enabling reliable LLM training and evaluation for clinical imaging. Strong data is the foundation of trustworthy radiology AI.

January 20, 2026

When AI Conversations Go Wrong and What High-Quality Data Can Do About It

A New York Times investigation shows how chatbot interactions can reinforce delusions and psychological harm. This response explains why the root cause is poor data quality and how Centaur.ai’s collective intelligence, performance-measured annotation, and superhuman datasets help teams build safer, more reliable high-stakes AI systems.