Join the movement and participate on Uniswap.
Participate
Retailers operate in one of the most competitive, low-margin industries in the world. They’re under growing pressure from both online competitors and evolving expectations from their customers and staff.
We believe there are three major challenges facing retailers today that could be addressed or mitigated through the use of AI:
For AI to be able to address these challenges, we must first teach AI to understand the physical environments and constraints of retail operations. Without spatial computing, the art of teaching digital things to understand physical space, AI remains incapable of helping retailers. This is further exacerbating the threat from online competitors that, currently, can better employ AI.
In this post we wish to demonstrate how collaborative spatial computing makes the physical world accessible to a new kind of spatial AI, allowing retailers to turn the tide and even gain new competitive advantages over online commerce.
It is estimated that retailers lose close to a trillion dollars in revenue annually to out-of-stock issues and poor shelf execution. Shelf space is a finite, high-stakes resource: Knowing where to place each product and how much shelf space to give them is one of the greatest challenges facing retailers, but also one of the greatest sources of potential revenue gain.
It is impossible to make data-driven decisions about shelf optimization without data on shelf performance. You can’t optimize the use of your shelf space without knowing where your products actually are in the store.
Attempting to solve this problem, retailers are resorting to planograms: aspirational maps of how products should be placed in their stores. However, planogram compliance is costly, difficult, and has its own inherent drawbacks: No two stores are alike, and one size does not fit all.
Globally, planogram compliance is low, meaning the data you get on how your shelves are performing is inaccurate and unreliable. In a way, the main stakeholder of the planogram is actually the analytics department. They need to know where products are to be able to analyze how the store could be optimized. The problem is that planograms only tell them where products should be, which often does not reflect the reality on the ground.
As such, retailers with low planogram compliance struggle to make any kind of data-driven decision. Garbage data leads to garbage insights.
The retailers that have high planogram compliance are instead locked into incredibly rigid and costly workflows that typically limit the amount of changes and experiments they can run. Many of the world’s leading retailers can only make 1-2 changes to a shelf’s layout per year.
But product placement is only one of the difficulties with managing your shelves. Once you know where to place your products, you still have to make sure that shelves have not run out of stock, and that the presentation of the product and shelf is conducive to conversion.
Making sure that products are at the front of the shelf is an important daily task for workers on the floor, because it increases product visibility and makes products more attractive to customers. This is a constant battle between entropy and information. It’s difficult to know when products need to be fronted or restocked.
Some retailers are experimenting with hundreds of shelf-mounted cameras to detect out-of-stocks and presentation issues, but have struggled with getting ROI. The margins in retail are so low, and the cost of implementing and maintaining these systems is so high, that retailers have largely given up on these kinds of static camera solutions, instead turning their hopeful eyes towards the near future of robotics.
Spatial computing allows us to capture the reality on the ground and produce a centimeter-accurate map of true product placement using handheld devices, wearables or robots. This gives retailers the ability to generate accurate and actionable heat maps, showing them the real-time impact of product placement experiments.
We believe that this will spell the death of the planogram. Our early users are telling us that this technology is letting them iterate faster, run more experiments, and optimize their store better for their unique demographics and layout.
But better analytics is just one of the many benefits of capturing the reality on the ground. Since our spatial AI knows the precise 3D location of where a picture was taken, we can go beyond detecting issues in screen space and correctly place them in world space. So, instead of mounting hundreds to thousands of battery-driven mini-cameras on your shelves, we see a future where smart glasses passively detect and report issues on the shelf continuously.
You can already use your phone and robots, today, to interact with our spatial AI, but having proactive smart glasses is not far away. Through our collaboration with Mentra, we feel confident that this feature can be shipped this year and many of the world’s largest retailers are already on the waitlist.
Retailers lose nearly a trillion dollars annually due to poor shelf execution and lack of accurate, real-time data needed to make informed decisions. Traditional planograms are unreliable, inflexible and difficult to enforce. Spatial computing offers an accurate alternative that empowers retailers to optimize shelves in real time, increase revenue, and reach better ROI than costly planogram compliance or shelf mounted cameras.
When a shopper is looking for a product, they often turn to a staff member. How often is the staffer able to help? Research indicates that 6% of shopper baskets would contain at least one more item if the staff themselves were better prepared. So why aren’t they?
Retail is one of the industries most affected by staff turnover. Globally, more than 50% of retail workers will change jobs any given year, and it takes up to a month for a new staff member to become fully proficient. Walmart employs 1.6 million associates in the US and spends around 100 million USD on first day salaries alone every single year. Maintaining a knowledge base and functioning store routines is incredibly difficult with such high turnover rates, a problem that is further exacerbated by low-skill and low-motivation workers that may struggle with the local language.
Spatial computing allows us to address these problems in two meaningful ways:
First, since we can map the precise location of products, we can provide step-by-step AR guidance to staffers complete with route optimization. This way, even a first day employee can find every single product and task location instantly with only minutes of training on how to use their phone to navigate the store.
We’ve demonstrated with clients that we can reduce walking distance by as much as 40% when collecting a basket of items. In addition, store associates, especially newer hires, tell us they spend at least 30 minutes per day looking for products even when they’re not doing click-and-collect.
Second, the ability to create notes and tasks in shared augmented reality helps the staffers communicate more effectively with each other. One six month pilot with a large supermarket found that this single feature saved them at least 15 minutes per day per employee on handovers alone.
Even more interesting is that this also helped them manage their low-skill workers better, especially those with mild cognitive impairment. They now report that their cognitively impaired staff operate at the level of their peers.
Soon, our next generation proactive AI will automatically create tasks for staff members and give them step by step guidance through their phones and glasses. We’ve already demonstrated that our spatial AI can identify restocking tasks, cleaning tasks and product presentation tasks.
Summarizing this, spatial AI makes it possible to improve worker productivity and reduce the cost of training. Very soon, each store associate will have an AI copilot. Smart glasses and other wearables will replace static cameras in the store, and spatial AI will autonomously detect, report and delegate tasks to keep the stores running and the shelves stocked.
This is the world of hybrid robotics, where human workers are augmented with artificial intelligence, allowing for the perfect combination of AI cognition with human locomotion and world manipulation. This technology promises to turn low-skill workers into expert contributors fast.
High staff turnover and low training levels in retail lead to inefficiencies, with even simple tasks like finding products taking up significant time. Spatial AI and augmented reality can dramatically boost productivity by guiding staff with real-time navigation, shared task systems, and AI-generated instructions. The future of retail is hybrid robotics: high-performing AI augmented workers.
eCommerce is growing nearly three times faster than traditional retail, not just because of delivery speed or convenience, but because it’s smarter. Every interaction feeds an algorithm that improves search, enhances discovery, and enables real-time personalization. Physical retail, by contrast, remains mostly static, unable to adapt to individual intent or capture the behavioral insights that online platforms rely on.
We believe the key advantages of eCommerce can be boiled down to three functions:
In-store, shoppers have only two options for finding a product: brute-force wandering or asking for help (something that nearly 70% of shoppers prefer not to do). As a result, customers leave frustrated and baskets shrink. According to a study by Oriient, up to 40% of shoppers fail to find at least one item on their shopping list during a visit and in half of those cases, the product was in stock but simply overlooked.
But the absence of search functionality doesn’t just reduce sales, it limits strategic decision-making. In eCommerce, every search is a signal: not only of what sells, but of what’s wanted but unavailable. Retailers can optimize stock, pricing, and merchandising based on that demand data. Physical retailers, on the other hand, see only what was sold and remain blind to what was searched for but never found.
Spatial AI fixes this.
By making the store machine-readable and context-aware, we transform the customer journey from analog to adaptive. Shoppers can use app-free AR navigation to instantly locate products and receive AI-generated recommendations right on their phones, without needing to install anything. Every interaction becomes a data point that feeds back into retail operations, merchandising, and store layout decisions.
This doesn’t just improve search, it unlocks contextual discovery and personalized experiences. Retailers can reclaim shopper attention from competing apps and deliver promotions, content, or suggestions that are directly tied to a shopper’s real-time location and behavior.
Importantly, this technology also makes retail more accessible. Today, we already support blind and partially sighted shoppers with spatially aware audio guidance, enabling them to independently navigate stores and locate products. We’re also partnering with leading electric wheelchair manufacturers to enable autonomous, route-optimized navigation in-store. And soon, the same infrastructure will power robotic personal shoppers capable of navigating and fulfilling errands on a customer’s behalf.
Physical retail lacks the intelligent search, discovery, and personalization that make eCommerce so effective, leading to missed sales and poor shopper experiences. Spatial AI adds a digital layer to physical stores, enabling app-free AR navigation, real-time recommendations, and gives retailers the data they have been missing about customer interactions.
The future of retail isn’t about imitating eCommerce, it’s about unlocking the strengths that only physical retail can offer: real-world presence, immediate gratification, and human connection. But for too long, those strengths have been held back by poor data, inefficient workflows, and outdated tools. With spatial AI, physical stores can finally become intelligent environments: adaptive, data-rich, and responsive to both shoppers and staff. By making stores machine-readable, we unlock better shelf execution, smarter workflows, and personalized customer journeys.
Spatial AI will augment both workers and shoppers with an AI-powered advantage. Retail is trying to solve trillion dollar problems and making the physical world accessible to AI through collaborative spatial computing will be central to that.
Auki is building the Auki network, a decentralized machine perception network for the next 100 billion people, devices and AI on Earth and beyond. The Auki network is a posemesh, an external and collaborative sense of space that machines and AI can use to understand the physical world.
Our mission is to improve civilization’s intercognitive capacity; our ability to think, experience and solve problems together with each other and AI. The greatest way to extend human reach is to collaborate with others. We are building consciousness-expanding technology to reduce the friction of communication and bridge minds.
X | LinkedIn | YouTube | AukiLabs.com | Medium
The Auki network is a posemesh: a decentralized machine perception network and collaborative spatial computing protocol, designed to allow digital devices to securely and privately exchange spatial data and computing power to form a shared understanding of the physical world.
The Auki network is an open-source protocol that powers a decentralized, blockchain-based spatial computing network. Designed for a future where spatial computing is both collaborative and privacy-preserving, it limits any organization's surveillance capabilities and encourages sovereign ownership of private maps of personal and public spaces.
The decentralization also offers a competitive advantage, especially in shared spatial computing sessions, AR for example, where low latency is crucial.
X | Discord | Telegram | Whitepaper | DePIN Network
