platoseedplatoseed
Proof of Human - invisible human verification
Roundtable Proof of Human is an API that invisibly detects bots and verifies human identity. Founded by 2 Princeton cognitive scientists, Roundtable leverages world-class research with enterprise-grade accuracy and latency so websites can verify their human users friction-free.
Roundtable offers proof-of-human verification that runs in the background to prevent fraud and ensure data quality without disruptive CAPTCHAs. It uses behavior analysis and research-grade AI to detect bots and suspicious activity in real time, with privacy-preserving design.
Roundtable provides an API for invisible human verification that analyzes user behavior (typing, cursor movement, scrolling, and other interaction patterns) in real time to identify real users versus bots or AI agents. It is integrated with minutes-long setup, operates without collecting PII, and offers explainable session reports. The service scales from day one to millions of sessions and promises near-instant API latency with an SLA.
Who it’s for: Industries requiring trusted respondent integrity and fraud prevention across user interactions, including market research, financial services, and ticketing for live events.
mention of enterprise customers and scalable deployment, demo booking, and public research-style reports; indicates traction and ongoing product development
CEO of Roundtable. Builds Proof of Human, an invisible Turing Test for the Internet. Before Roundtable, Mayank completed a PhD at Princeton University working at the intersection of cognitive science and artificial intelligence, publishing in high-impact venues including Science, PNAS, and Psychological Review.
Matt is a co-founder of Roundtable, a company using behavioral metrics for invisible fraud prevention. Before Roundtable, Matt was a PhD student at Princeton University, where he used machine learning and behavioral and experiments to study how people make decisions. Before his PhD, Matt majored in both statistics and economics at the University of Toronto, and worked in data science and econometrics.
AI agents reliable enough to do real work
Enterprise Deepfake Detection
