Teradyne vs. JAKA, NSF’s $30M Robot Center and Carbon’s 150M Plants

Three robotics stories reveal where the industry’s next competitive battles are moving: patents, people and data.

Teradyne Robotics filed a patent infringement case against JAKA at Europe’s Unified Patent Court. The National Science Foundation committed $30 million over five years to a center studying how people and robots adapt to one another. Carbon Robotics says its Large Plant Model was trained on roughly 150 million labeled plants.

In Episode 2 of The Business of Automation, Aaron Prather, Director of Market Intelligence at A3, connects these announcements to a larger shift. More of robotics’ value and risk now sit outside the machine itself.

Teradyne alleges that multiple JAKA collaborative robot models infringe patents covering hardware and software associated with Universal Robots. Those allegations have not been proven in court.

The larger business story is where the case was filed. Historically, enforcing patent rights across Europe could require separate proceedings in multiple countries. A ruling from the Unified Patent Court could potentially affect JAKA products across 17 of its 18 current member states.

For robotics manufacturers, integrators and buyers, intellectual property exposure is becoming an increasingly important part of evaluating an automation purchase.

The second story focuses on the Center for Human and Robot Co-Adaptation, led by the University of Texas at Austin. The center brings together researchers from six universities and industry partners including Amazon, Apptronik, Diligent Robotics, Google DeepMind, Hello Robot, MassRobotics, NVIDIA and Robust AI.

The research is not only about making robots more capable. It is about understanding how people and robots learn from one another and adjust their behavior over time in homes, hospitals, workplaces and public spaces.

A robot can perform every task correctly and still fail if the people around it do not trust or accept it. Aaron explains why training, workforce design and clear team leadership should be part of an automation strategy from the beginning.
Carbon Robotics provides the clearest example of value moving away from the robot itself. Its LaserWeeder uses computer vision and lasers to identify and eliminate weeds, but its deeper competitive advantage may be the data behind the machine.

Carbon says its Large Plant Model was trained using roughly 150 million labeled plants collected across more than 100 crops in 15 countries. That scale allows growers to adapt the system to field conditions using only a few images instead of waiting for an entirely new crop-specific model.

The relationship also introduces another business question. iMerit helped provide the annotation infrastructure supporting Carbon’s data flywheel, and iMerit has now been acquired by EXL. When an outside vendor becomes critical to a robot’s performance, a change in ownership can become part of the customer’s risk.

Together, these stories raise three questions every automation buyer should ask: What is the intellectual property risk? How will people respond to the system? Who owns the data the machine generates?

Read the deeper analysis, with every source linked and every number cited:
thebusinessofautomation.substack.com
This episode is presented by the A3 International Robot Safety Conference, November 3-5, 2026, at Huntington Place in Detroit.

Learn more and register:
automate.org/events/international-robot-safety…

Follow Aaron Prather:
linkedin.com/in/amprather

Explore A3:
automate.org

businessofautomation.fm