Honda funds a Cal State Fullerton course to teach engineers to trust robots, not just operate them

By CSUF News · September 28, 2026.

Honda has committed nearly $100,000 to Cal State Fullerton (CSUF) to fund "Building Trust in Human–AI Robotics Teaming for the Future Manufacturing Workforce", a one-year project led by Bingling Huang, assistant professor of mechanical engineering and director of the university's Design for Intelligent Systems and Learning Laboratory. Twenty students were selected this summer to take part in the project; over the course of the academic year, more than 200 engineering students will receive this material embedded in existing courses, through demonstrations, simulations and hands-on projects.

The core of the project is concrete: students will build a virtual manufacturing environment with simulated robots, assign them tasks and observe how they respond, for both research and teaching purposes. The list of tasks AI can already take on in a plant is long, as the article details: recognizing parts, guiding robotic arms during assembly, inspecting products for defects, anticipating when equipment will need maintenance and optimizing production scheduling.

The account of graduate student Rahul Pravinbhai Lunagariya, a research assistant on the project, illustrates the problem Honda wants to tackle. Before arriving at CSUF, he worked at one of India's largest salt producers, with conveyor belt systems and heavy automation to move coal and lift industrial components such as shafts and boiler parts, and later as a production engineer at an auto parts factory. His conclusion, quoted in the original article: "workers either did not trust the automation systems around them, or the company avoided automating altogether", and the cause was "a lack of understanding and transparency about how those systems behaved": if the machine is unpredictable, the operator becomes reluctant or resistant.

That is the reading that interests us. The project does not teach how to program smarter robots, but how to diagnose whether an autonomous system is "understandable, reliable and appropriate for human collaboration", in Huang's words. It is a subtle but significant shift: it treats trust not as a desirable side effect of automation, but as a technical skill that can be defined, measured and designed, on a par with a sensor's accuracy or a robotic arm's speed. That Honda is the one paying for it —a manufacturer with decades of its own industrial robotics— suggests the industry already knows, from shop-floor experience, that automation's bottleneck is not only technical: it is a matter of adoption, and adoption depends on whether the operator understands what the machine is going to do next.

It is worth not overstating the fact: nearly $100,000, one campus and a one-year run, the kind of university-industry agreement signed every semester at dozens of engineering schools in the United States. There is no technical breakthrough to report here, but rather a signal of where the training of engineers who will work side by side with autonomous systems is heading. And that signal fits something we have been observing in our coverage of AI's impact on employment sector by sector: in manufacturing, as in healthcare or banking, what survives automation is not the routine task but judgment —here, the judgment to decide whether a machine deserves trust— and that judgment has to be taught explicitly, because it does not emerge simply from using the tool.

In the long term, the goal Huang states —engineers who know how to use AI and also how to evaluate it— is exactly the kind of capability needed for automation to stop being a source of friction on the plant floor and start freeing workers from the heaviest and most dangerous tasks, such as lifting shafts or boiler parts, toward supervisory and design roles. It is not a promise this project will fulfill on its own, but it points in the right direction: the abundance we defend as a horizon does not arrive by decree; it arrives when someone sits down to design, course by course, how a human and a machine learn to trust each other.