Science you can actually trust
An AI scientist that reads the world’s research, weighs the evidence, and shows its work, so anyone can tell what’s actually true.
SciTrue started in 2023 with one question: what if anyone could check a claim against the real research in seconds, and see exactly which papers, sentences and evidence stand behind the answer? Not a black-box verdict, but the reasoning shown in full.
Today SciTrue reads across more than 300 million papers to weigh the evidence for and against any claim, surface the assumptions and nuances, rate the credibility of every source, and let you chat directly with any paper. It’s the second opinion we all wish we had when a confident voice tells us what the science “says.”
Where we’re going
Claim verification is only the beginning. We’re building toward a Science OS: an AI scientist that doesn’t just verify what’s known, but reads the research, designs and applies experiments, draws inferences, and runs deep analyses on the data to help extend the frontier of what we can discover. It will be a trustworthy research partner for everyone, from the curious reader to the working scientist. The science that shapes our world should be transparent, verifiable, and open to all.
The people behind SciTrue
The foundations of SciTrue were laid during his PhD research, where he began developing natural language algorithms designed to prevent scientific misinformation and enable automated reasoning. Building on these early ideas, Neşet founded SciTrue to create intelligent systems that make scientific knowledge transparent, verifiable, and accessible.
Joined SciTrue in 2025. A Master of Science student at the University of Auckland specialising in trustworthy AI, he holds a Bachelor of Engineering and a Postgraduate Diploma in Computer Science. With experience across front-end, back-end, and cloud infrastructure, he brings full-stack engineering to SciTrue.
Principal Research Scientist at Microsoft Research (previously Lead Research Scientist at the Allen Institute for AI), working on grounding and feedback-guided reasoning in large language models.
Professor of Computer Science at the University of Auckland and an internationally recognised AI expert (previously IBM, Cycorp, Carnegie Mellon), specialising in automated reasoning and knowledge acquisition.
A leading expert in eScience and data-driven scientific discovery, focused on reproducibility, knowledge representation, and scalable data systems for research.