Discover our mission and the innovative work we do in advancing artificial intelligence
Watari Labs is an advanced artificial intelligence laboratory dedicated to exploring, experimenting, and innovating across the full spectrum of AI research. We believe that the future of technology lies in continuous experimentation testing, iterating, and refining models that can redefine how humans and machines interact.
We operate as a living laboratory where ideas are transformed into real-world applications through rigorous research and experimentation. Our work spans across key AI domains including Reinforcement Learning (RL), Deep Learning (DL), Machine Learning (ML), and Large Language Models (LLMs).
We strive to push the boundaries of what AI can achieve by building practical, scalable solutions that can be tested, evaluated, and deployed in real-world scenarios. Every experiment brings us closer to unlocking the full potential of artificial intelligence.
Cutting-edge experiments across multiple AI disciplines, pushing the boundaries of what's possible
The next evolution in AI evaluation: Watch as language models engage in intelligent debates, analyze arguments, and demonstrate critical thinking across multiple topics
AI models will engage in debates across politics, science, philosophy, technology, and more, testing their reasoning abilities and argumentation skills.
Models will be judged on their ability to identify logical fallacies, evaluate evidence quality, and construct compelling counter-arguments.
Experience how AI models perform in structured debates, providing persuasive arguments and responding to opponent viewpoints.
Should AI development be regulated by governments?
Find answers to the most frequently asked questions about our AI evaluation platform and research methodology.
Get in touch with our research team. We're always excited to discuss AI innovations and potential collaborations.