Intrinsic, the Google-owned robotics company, used this week's Automate 2026 event in Chicago to preview a modular robot workcell designed to bring artificial intelligence deeper into manufacturing. The company says the new system is intended to help factories and machine shops build more flexible production lines that can be reconfigured quickly as products and processes change.
The centerpiece of the announcement is the Intrinsic Intelligence Cell, a reference design for a robot workcell built around IntrinsicOS. The company describes it as a skill-based automation system aimed at complex assembly tasks. According to Intrinsic, the setup is meant to support fast tool and process changes, work with different hardware and software combinations, and use AI to make robotic systems more adaptive.
Intrinsic says the concept is aimed at businesses ranging from smaller machine shops to large manufacturers such as Foxconn, which is listed as a factory of the future partner. A custom version of the workcell for electronics assembly is expected to be piloted in Foxconn facilities later this year. At Automate 2026, Intrinsic is showing a version of the system that uses a FANUC robot to demonstrate electronics assembly tasks. The company says the collaboration with FANUC is part of a broader effort to improve hardware interoperability on its platform.
The workcell is positioned as a software-first approach to industrial robotics. Intrinsic says that by combining AI capabilities such as perception, automated motion planning, and grasping and insertion skills, the technology could reduce the need for traditional robot programming. The company argues that this could make advanced automation more accessible to machine operators and system integrators, while also supporting high-mix production runs and greater output from existing equipment.
Wendy Tan White, chief executive of Intrinsic, said the combination of AI and modular production could reshape operating models across manufacturing. She said greater flexibility in both hardware and software may change the economics of production and open new ways for manufacturers of different sizes to build.
Intrinsic also used the event to update its global AI for Industry Challenge, a competition launched with Open Robotics to tackle one of electronics assembly's more difficult problems, the manipulation of cables and connectors. The company says the challenge carries a $180,000 prize pool and has drawn strong international interest. More than 5,000 people have registered across 1,600 teams from over 115 countries.
Intrinsic said eight teams have reached near-perfect scores in the simulation phase so far. Of the participants, 93% report proficiency in Python, 73% have ROS experience, and 47% work in AI, machine learning, or software engineering. Only 14% come from robotics backgrounds, which Intrinsic says suggests the problem is attracting talent from beyond the traditional robotics field.
Thirty-one teams have advanced to the next phase, where they will test their solutions using Intrinsic's software tools, including Flowstate and the Intrinsic Vision Model. The top 10 finalists will then remotely deploy their solutions to a physical industrial workcell at Intrinsic's California headquarters for real-world testing.
Intrinsic says the broader goal is to make Physical AI more practical and affordable for manufacturers, developers, and systems integrators. The company is targeting a large community of ROS developers as well as industrial partners who want to move faster from prototype to production. In its view, modular robotics paired with AI could widen access to automation and lower the barrier to entry for factories and machine shops that want to adopt more advanced systems.