HD Hyundai Robotics says Palantir AI cut robot troubleshooting from 21.5 days to 7.9
The company reports faster problem-solving but still treats programming-cost savings as a forecast. Its workflow keeps engineers reviewing and refining AI suggestions.
Listen to this story
The audio brief
Story brief
3 key pointsHD Hyundai Robotics is shifting robot-program troubleshooting from repeated manual adjustments toward a human-reviewed workflow: Palantir-built agents inspect existing programs and propose changes, while engineers validate them. Since deployment in August, the company reports average resolution time fell from 21.5 to 7.9 days, a 63% decrease—an operational result, not an AI benchmark. The collaboration could also shift engineers toward identifying problems and developing AI solutions, but the projected programming-cost reduction of more than 50% remains unachieved, and further platform-linked applications are still plans.
- 01
The previous process required engineers to repeatedly adjust robot operating sequences, routes, and speeds based on their experience.
- 02
HD Hyundai Robotics disclosed the results on November 1 at Palantir’s Sovereignty Bootcamp in New York.
- 03
Executives from both companies plan regular steering committee meetings to set priorities and explore additional applications.
Robot troubleshooting that once took HD Hyundai Robotics an average of three weeks now takes just over a week, the company says. On November 1, it disclosed that AI agents developed with Palantir had reduced average problem-solving time from 21.5 days to 7.9 days since their deployment in August—a reported cut of 63%.
The results were presented at Palantir’s Sovereignty Bootcamp in New York, where HD Hyundai Robotics Chief Technology Officer Ahn Seong-hwan outlined the collaboration and its industrial AI strategy. The work combines AI, simulation and engineering expertise to improve how robots are programmed.
From repeated adjustments to AI suggestions
Previously, engineers manually checked robot movements and repeatedly adjusted operating sequences, routes and speeds. That work depended heavily on skilled engineers’ experience.
In the new workflow, AI agents analyze existing robot programs and suggest improvements. Engineers then use their field expertise to review and refine those recommendations.
Palantir Chief Revenue Officer Ryan Taylor said the companies initially focused on projects intended to produce measurable results within a short period. The troubleshooting figures describe one such operational measure, rather than a score for the agents on a standalone AI test.
Company-reported average troubleshooting time
HD Hyundai Robotics attributes the reduction to AI agents deployed in its problem-solving processes since August.
Faster troubleshooting is not yet a cost result
Alongside the timing results, HD Hyundai Robotics expects the technology to reduce robot-programming costs by more than 50%. That is a forecast, not a cost reduction it has reported achieving.
A broader assignment for field engineers
HD Hyundai Robotics also plans to broaden engineers’ responsibilities beyond robot programming and installation. It intends to train them to identify operational challenges and develop AI-powered solutions.
Senior executives from both companies plan regular steering committee meetings to set priorities and explore further applications. Taylor also described an intention to seek business opportunities by connecting HD Hyundai Robotics’ robots with Palantir’s Artificial Intelligence Platform. Those plans extend the same collaboration, but remain future applications rather than additional demonstrated results.
Sources
- koreaherald.comHD Hyundai Robotics cuts troubleshooting time 63% with Palantir AI
- upi.comSouth Korea's HD Hyundai deepens physical AI ties with Palantir - upi
Reader comments
Newest comments first. Replies stay oldest first.