About us
We aspire to be the standard by which human-AI decision-making is measured
Coalescent Cognition measures the performance of Human–AI teams. Drawing on nearly two decades of UCL research into human and AI metacognition, our methodology evaluates the human, the AI, and the interaction between them against a shared rational baseline.
The science is ready
Computational models of confidence, belief updating, and metacognition have matured over the past decade to where they can be deployed outside the lab.
Deployment is happening
Vertical AI in legal, finance, and clinical settings has moved from pilot to production inside major institutions. The teams using these tools are now making consequential decisions with them daily.
Accountability has arrived
Professional-indemnity underwriters are now asking what AI is used, how it is overseen, and what controls exist. The duty to verify LLM output is non-delegable.
Contact
Get in touch
For any inquiries please reach out to us below.