DARTS is an open research initiative working toward AI that genuinely understands its environment, builds its own model of the world, and generates new knowledge rather than recombining existing patterns.
All of our work is developed in the open.
World Models — AI that builds and navigates internal representations of reality, so it can imagine, plan, and act beyond what it has directly observed.
Causal Learning — Moving beyond correlation toward systems that understand why things happen, through exploration and interaction with the world.
Multi-Agent Systems — AI systems that communicate, coordinate, and learn from each other, because emergent learning through interaction is fundamental to how genuine understanding develops.
Contributions are welcome: research, code, writing, or simply spreading the work.
Reach out at contact@dartsinitiative.com
