The Robotics Advantage Is Now About The Ecosystem

John Healy, Vice President Client Computing and Physical AI Group, General Manager Industrial and Robotics Division at Intel Corporation.

getty​As robotics shifts from fixed-function machines to AI-defined systems, one question becomes hard to ignore: Are we entering a new robotics era, or are we still trapped in the architectures of the past?

​For years, robots have captured our imagination as machines that lift, deliver, assemble, fly and walk. Today, the real revolution may not be in how they move but in how they are built, connected and continuously improved. ​

Many automation architectures were designed for a more predictable time, as a traditional robot performs a defined task inside a tightly controlled system.

An AI-enabled robot must absorb new models, sensors and workflows; collaborate with other machines; and improve without requiring the entire deployment to be rebuilt.

Closed, vertically integrated systems may deliver near-term optimization, but they can also create high switching costs, slow integration and dependence on one supplier’s roadmap.

The emerging model is more pragmatic: open at the foundation and proprietary at the edges. Similar to the internet, where common protocols became shared infrastructure while businesses differentiated above them, robotics is beginning to separate foundational layers from competitive ones.

Robotics adoption is accelerating. The International Federation of Robotics reported “542,000 industrial robots installed in 2024—more than double the number 10 years ago,” with annual installations projected to surpass 700,000 by 2028.

Growth is increasingly extending beyond the factory floor to autonomous mobile robots (AMRs), service robots and the first generation of general-purpose humanoids, all built as software-defined systems from the outset.

The rise of embodied AI is accelerating this transition. As robots become software-defined systems powered by rapidly evolving AI models, the ability to integrate, upgrade and coordinate across platforms becomes increasingly important. As fleets grow larger and more heterogeneous, common interfaces become economically essential.

Open robotics is emerging as the shared foundation for this new era. By providing reusable components such as middleware, simulation, hardware abstraction and fleet coordination, it allows developers to focus on innovation rather than rebuilding core infrastructure.

In 2025, the Open Source Robotics Foundation reported more than 984 million Robot Operating System (ROS) package downloads, up 85% year over year, with ROS 2 accounting for over 91% of all downloads. Governance has matured alongside adoption through the Open Source Robotics Alliance, which helps ensure the long-term stability of ROS, Gazebo and Open-RMF.

In other words, open robotics has evolved from a research curiosity to become industry infrastructure.

Consider a logistics operator running AMRs from multiple vendors. A decade ago, that meant separate systems and integration projects. Today, operators increasingly standardize on an open coordination layer, allowing diverse fleets to be managed through a single system while vendors compete on performance, reliability and operational outcomes.

This procurement dynamic would have been impossible in a closed world. Now, the operator can add additional vendors without ripping out its orchestration software, negotiate on outcomes rather than lock-in and upgrade the AI models independently of the hardware.

The debate is often framed as open versus closed, but successful robotics solutions will need both.

Openness expands access to talent, accelerates development and reduces dependency on a single vendor. Proprietary capabilities provide the performance, accountability, safety and domain expertise required in production.

With this model, leaders should standardize the layers customers expect to work across vendors: middleware, hardware interfaces, communications and APIs.

Differentiation will focus on outcomes: real-time performance, safety, AI optimization, domain-specific workflows, certification and lifecycle management. This is where companies build defensible value even as the foundation becomes more shared.

The economics change as a result. Hardware remains essential, but economic value shifts toward software, orchestration, AI, services and lifecycle optimization. As interoperability becomes expected, differentiation moves higher up the stack.

It would be inaccurate to assume the open model succeeds by default.

Some of the most valuable technology companies of the past two decades built their advantage precisely by owning the full stack. Vertical integration can deliver performance, reliability and a coherent experience that loosely coupled systems struggle to match. It also offers something enterprises value highly: a single accountable party.

Openness carries costs, including ecosystem fragmentation, integration complexity and broader security exposure. “Open” projects do not guarantee neutrality. Governance can ultimately reflect the interests of the vendor driving most of the development. And when an open component and a proprietary controller interact and something fails, liability can become genuinely murky.

These are reasons to be deliberate about where the boundary sits, who is accountable at each layer and how safety and security are governed.

• Can the system integrate with machines from other suppliers through open interfaces?

• Can hardware, software, sensors and AI models evolve independently?

• Who is responsible for safety, security and support when open and proprietary components interact?

• What data and interfaces remain accessible if the commercial relationship changes or ends?

• Are critical interfaces governed by open standards?

The answers reveal whether an automation investment is a flexible operating asset or the beginning of an expensive dependency.

As open platforms mature, robotics increasingly resembles a shared digital layer across industries. Manufacturing becomes more flexible and software-defined. Logistics and warehousing run heterogeneous robot fleets that integrate seamlessly. Healthcare, service robotics, autonomous vehicles and drones move deeper into everyday deployment pipelines. ​​

Meanwhile, as embodied AI and general-purpose humanoids move to scale deployment, the pressure to get this architecture right will only intensify.

The robotics revolution will be shaped by adaptable ecosystems where machines, models and partners can interoperate and evolve together. Open platforms are gaining ground because modern robotics demands a level of agility, interoperability and innovation that no single organization can deliver alone.

However, openness itself is not the strategy. The strategic choice is deciding which layers should become shared infrastructure and where differentiated value should be created. This distinction will define the next decade of robotics.​

Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?