Robots as Infrastructure: The Path to $8.6 Billion ARR

Coming to you live from San Francisco, this week's community update zeroes in on the exact mathematical and strategic roadmap that has Silicon Valley robotics companies realizing Auki's approach is entirely unique.

We discussed the concept of "robots as infrastructure," the power of co-embodiment, and why solving perception first is the ultimate cheat code for scaling physical AI.

The 1,000x Difference: Perception vs. Manipulation

Nils recently sat down with the CTO of a well-funded European robotics company focused on manipulation tasks (building robot hands, arms, and manipulation AI). When asked for their most optimistic timeline to reach $100 million in annual revenue, the answer was 2030.

We believe there is a path that is 10 times easier and 100 times bigger.

Robotics tasks generally fall into three categories: Locomotion, Manipulation, and Perception. Most of the industry is attempting to solve manipulation first—an incredibly fragile, high-risk endeavor. Auki is solving perception.

Take agriculture as an example. An Auki ecosystem partner is deploying "Mars rovers for vineyards." Rather than trying to solve the impossibly complex task of robotic grape harvesting, they deploy perception robots to drive the rows and detect crop-destroying diseases early. The yield increases immediately, paying for the robot on day one, all without a single mechanical arm.

The Back-of-the-Napkin Math

Applying this perception-first model to retail reveals the true scale of the real world web.

• Robots: If we deploy 100,000 perception robots by 2030 (acting as autonomous store managers generating tasks) at roughly $110/day, that generates $4 billion ARR.
• Smart Glasses: If we deploy 1 million Mentra smart glasses to retail floor workers at just $11/day, that adds another $4 billion ARR.
• Phones: The Cactus companion app is already generating $500/month per location. At 100,000 locations, that is another $600 million ARR.

By focusing purely on perception—understanding the space, the inventory, and generating actionable intelligence—we can scale an $8.6 billion recurring revenue business before the rest of the industry has figured out how to reliably fold laundry.

Robots as Infrastructure & Co-Embodiment

Once a perception robot is deployed in a store, it has captured that physical real estate. If that robot happens to have arms (a semi-humanoid), it becomes infrastructure.

Instead of other AI companies trying to sell and deploy their own robots into that same store, we enable co-embodiment. Third-party manipulation AIs can be deployed onto our existing hardware infrastructure. If we charge a simple software licensing fee to run third-party manipulation models on our fleet, the revenue compounds with near-zero deployment friction.

We are building the two most important pieces of infrastructure for the robotic age: the real world web (which collapses deployment time to under a minute), and the deployed robotic fleet itself.

Internal Demos: AI Insights and Shelf Version Control

Back in the lab, our internal demo days are bringing this vision to life.

• Cactus Insights: The AI now actively monitors the 3D domain and sales data. It proactively alerts managers to anomalies, such as, "This item in the cookie section is generating 3% of your total revenue. Consider moving it to eye level."
• Shelf Version Control: We have built a drag-and-drop shelf configurator that acts like GitHub for physical shelves. Headquarters can design the perfect layout, and the AI will generate the exact, step-by-step tasks required for floor staff to make reality match the plan.
• App-Free AR on Android: We previewed native, app-free WebXR navigation running directly in the Chrome browser on Android, matching our existing App Clip functionality on iOS.

Our strategy has surprised the industry. "We came out of left field, surprised everyone because we realized robots need the real world web and the smartest go-to-market strategy is perception first, and then co-embodiment."

If you want to hear the off-the-record updates and ask questions, join our weekly AMAs on Discord.

Watch the whole update on X.

About Auki

Auki is making the physical world accessible to AI by building the real world web: a way for robots and digital devices like smart glasses and phones to browse, navigate, and search physical locations.

70% of the world economy is still tied to physical locations and labor, so making the physical world accessible to AI represents a 3X increase in the TAM of AI in general. Auki's goal is to become the decentralized nervous system of AI in the physical world, providing collaborative spatial reasoning for the next 100bn devices on Earth and beyond.

X | Discord | LinkedIn | YouTube | Whitepaper | auki.com