humans& is a frontier AI lab building AI systems designed to strengthen how people work together — not to replace human labor, but to improve human communication, collaboration, and coordination. Founded in September 2025 by five co-founders with pedigree from xAI, Anthropic, Google DeepMind, OpenAI, Meta, and Stanford, the company believes the next frontier of AI requires deliberate innovations in long-horizon memory, multi-agent reinforcement learning, and user understanding — capabilities that won't "simply fall out of scaling up training." The company is building a full product stack (web + mobile) using Convex, Next.js, and React Native, targeting teams and organizations as the likely buyers. The thesis is a philosophical bet against the dominant autonomous-agent paradigm: rather than AI that works independently, humans& wants AI that makes groups of humans smarter together. They have also committed to open-source contributions and academic collaboration.