68% of auto companies use agentic AI, but only 5% have mature governance

what happened

GlobeNewswire, citing the Capgemini Research Institute, reports that 33% of automotive organizations already use agentic AI in product development and another 35% are running pilots. Adoption in non-core engineering work is expected to rise from 26% now to 68% over the next two to three years. At the same time, 83% of executives say the technology could give engineers more time for core engineering instead of administrative and coordination tasks.

The study surveyed 800 senior automotive executives in 12 countries across North America, Europe and Asia-Pacific. Companies capturing the most value are achieving gains in time-to-market at up to three times the pace of lower-performing peers. They report development cycles that are 17% faster, a 19% improvement in engineering productivity and a 19% reduction in time-to-market. Yet only 5% describe their AI governance as fully mature. Confidence in agentic AI reaches 59% in research and concept work but falls to 40% during development.

The US and Europe show the strongest strategic execution: 53% and 46% of OEM respondents, respectively, are implementing AI initiatives, while 18% and 16% have a comprehensive cross-functional strategy, vision and roadmap. In China, four in ten respondents are implementing AI, but only 7% report comparable strategic maturity. Nearly two-thirds, or 64%, say vehicle data mainly informs future programs rather than improvements to vehicles already in use. Six in ten executives cite process inefficiencies as the main source of delays, 45% lack an AI strategy and roadmap, and only 30% have shared outcome metrics while 10% share lifecycle accountability across the ecosystem. The survey was conducted in July 2026 and supplemented by interviews with 24 industry executives and technology providers.

why it matters

For automotive markets, the findings suggest that returns from agentic AI will depend on data quality, traceability and governance as much as on deployment itself, with coordination across OEMs, suppliers and technology partners remaining a central constraint.

instruments