70% of the world’s GDP is still tied to physical labor The market for Physical AI, machines that can act in the real world, is 3x bigger than LLMs. Meet @broodsugar from @AukiLabs building the "Real World Web".

2/ The Internet made all our digital information browseable and searchable. Auki is doing the same for the physical world, making real-world spaces browsable, searchable, and navigable to robots. Auki is building a decentralized machine perception network - a shared spatial understanding that lets multiple robots and devices see and coordinate in the same space. Think of it as Google Maps + GPS, but for indoor environments. Videos recorded by smartphones or smart glasses inside a venue allow Auki to create a 3D map of that space so a robot can later navigate it without ever being there before. Crowdsourced perception.

3/ Neither Tesla nor Meta have maps of your home, office, or store interiors. Autonomous systems today can’t coordinate because they lack a shared reference frame. Auki’s Real-World Web aims to fix this by enabling machines to understand and coordinate within physical spaces. This isn’t about decentralizing data collection. It’s about decentralizing data ownership. Even Auki doesn’t need to host your 3D map. Venues can self-host their spatial data; preserving privacy while enabling interoperability.