We accelerate AI breakthroughs

Collaborating with leading AI teams to ensure quality at every step of the AI product cycle, from training to model evaluation

AI is a lot of work

As the technology becomes ubiquitous, a vast tasker underclass is emerging — and not going anywhere.

Our Story

Centaur Labs was born out of research at the MIT Center for Collective Intelligence, which explores how people and computers can be organized so that – collectively – they act more intelligently than any person, group or computer alone.

While our founder, Erik Duhaime, was working on his PhD research in collective intelligence, he was inspired by his wife’s experience in medical training. He started running experiments on how to best combine the opinions of multiple medical professionals and AI algorithms to most accurately assess skin lesions for cancer.

He found that the key is to trust opinions based on performance-based metrics – rather than years of experience or credentials – and to then identify how the opinions of different people and computers best complement each other. This methodology produced more accurate results than trusting even the best individual doctor, or the best AI algorithm alone. This research became the foundation of Centaur Labs.

What we do

We provide high-quality, trusted data annotation at scale, to power the development of superior AI models.

How we do it

We unlock the power of Collective Intelligence. By gathering and intelligently synthesizing multiple expert opinions, we achieve better-than-expert quality at scale.

The future is Centaur

Centaur Labs draws inspiration from centaur chess —a style of chess where human players collaborate with AI to optimize for high quality decision making.
‍
Just as centaur chess combines the strengths of humans and computers, our data labeling methodology combines the human judgment of a global network of subject matter experts, with complex quality measurement and AI.

The resulting Collective Intelligence enables our platform to deliver superior data quality to fuel the next generation of AI development.

Our expert advisors

From world-class researchers, to industry leaders, our advisors bring deep expertise in health and AI and help keep us on the cutting edge.

Thomas Malone, PhD

Founding Director, Center for Collective Intelligence

Thomas Malone is founding director of the MIT Center for Collective Intelligence, and author of "Superminds: The Surprising Power of People and Computers Thinking Together”. He cofounded 4 software companies and holds a PhD from Stanford.

Matthew Lungren, MD

Chief Medical Information Officer

Dr. Lungren is Chief Medical Information Officer at Microsoft+ Nuance and a clinical machine learning researcher with a part-time practice at UCSF. Previously, he led the Stanford Center for AI in Medicine and Imaging. He has over 150 publications on healthcare AI, including work on multi-modal data fusion, computer vision, and NLP.

Tina Kapur, PhD

Executive Director, National Center for Image Guided Therapy

Tina Kapur is the Executive Director of Image Guided Therapy at Brigham and Women's Hospital. Her research specializes in medical image computing and surgical navigation, with numerous publications and patents. Dr. Kapur earned her Ph.D. in Electrical Engineering and Computer Science from MIT.