.png)
.png)
Last week, our team attended Automate 2025, a premier conference for automation and robotics bringing together over 850 exhibitors and thousands of attendees. The event showcased the latest innovations in the field and sparked important conversations about where the industry is heading. Here are our key takeaways from the event.
The convergence creating a physical AI tipping point
A recurring theme throughout the conference was that we've reached a critical inflection point for “physical AI,” the use of artificial intelligence techniques to learn about, understand, and directly interact with the physical world. This isn't just about technological capability—it's about the convergence of multiple factors that make now the right time for widespread implementation:
- Hardware costs have significantly decreased, making advanced robotics and automation accessible to organizations of all sizes
- Edge computing capabilities have matured, enabling more sophisticated processing directly on devices
- Cloud infrastructure is robust and reliable for handling complex data workloads
- Hybrid architectures provide flexible deployment options, blending edge and cloud processing
Throughout various keynotes and panel discussions, industry leaders emphasized that this convergence represents trillions of dollars of opportunity across industries. The technology is now affordable, powerful, and practical enough to cross from theoretical possibility to business reality.
Software is a primary driver of business value in physical AI applications
No doubt, the booths drawing the most attendee visits and photos were those with humanoid robots showing off an impressive array of dexterity. While hardware innovations continue to impress, a surprising insight emerged consistently across presentations and conversations: software is now the primary driver of value in physical AI applications.
This represents a significant shift. Historically, robotics and automation were hardware-centric fields where physical capabilities defined what was possible. Today, with increasingly commoditized and affordable hardware platforms, the differentiating factor is the software that orchestrates hardware systems, analyzes their data, and enables them to adapt intelligently.
For example, our recent wine-pouring demonstration shows how the Viam platform makes it possible to deploy AI in real-world settings with speed and flexibility using a standard robotic arm like those already found in many factory floors.
That said, the resulting question for businesses isn't just "Can we automate this process?" but rather "Does it make business sense to automate this process, and how can we do it most effectively?" Just because AI can perform a task doesn't automatically mean it should—the business case must be sound, considering factors beyond technical feasibility.
Interoperability and standardization are essential
Perhaps the most pressing challenge highlighted at Automate 2025 was the need for interoperability and standardization across automation systems. Many organizations are hesitant to adopt new technologies not because of their cost or capabilities, but because of the integration complexity they introduce.
IT teams simply cannot support dozens of disconnected point solutions, each with its own management interface, security model, and maintenance requirements. This challenge is particularly acute in existing facilities with legacy technology investments that cannot simply be replaced.
One example of this is Viam’s partnership with The New York Islanders and UBS Arena to develop QuickQueue, a tool that directs fans to lines with the shortest wait times around the arena. The Viam platform plugs directly into the arena’s existing infrastructure of cameras and sensors, which made it easy to get started: we simply deployed our software on top of UBS Arena’s security cameras. The result is a user-friendly mobile application that gets fans back to their seats as soon as possible.
The industry is recognizing that, similar to how cloud computing evolved, robotics and automation need common frameworks that simplify deployment, management, and integration. Several presenters described the need for a universal platform that would make provisioning and managing physical resources as straightforward as spinning up cloud instances.
The importance of connective tissue across physical systems
Ultimately, what many organizations are seeking is the connective tissue that can orchestrate interactions across their diverse physical systems. This encompasses:
- Seamless integration between hardware and software components
- Centralized management and monitoring of distributed devices
- Consistent security models and access controls
- Standardized data collection and processing pipelines
- Simplified deployment and configuration workflows
The companies and presenters who demonstrated the most successful implementations had all addressed this critical need for integration and orchestration, whether for building digital twins, managing fleets of machines, or evolving the technology stack to track with the pace of innovation. They recognized that the value of automation isn't just in individual capabilities but in how those capabilities work together as part of a cohesive system.
Looking forward
Automate 2025 made it clear that we're entering a new era for physical AI and automation—one where the technology is mature enough to deliver substantial business value across industries. The most successful implementations will be those that balance technical capability with business pragmatism, integrate seamlessly with existing systems, and provide the flexibility to evolve as needs change.
As organizations navigate this landscape, they need partners and platforms that can simplify the complexity, bridge the gap between hardware and software, and provide the connective tissue that brings diverse technologies together. At Viam, that's precisely the challenge we're working to solve.
