According to Capgemini’s latest research, nearly 70% of automotive organizations are already using or testing agentic AI in product development.
Agentic AI is no longer just a pilot project in the automotive world, it is rapidly creating a massive divide between industry leaders and those struggling to keep up.
According to Capgemini’s latest research, nearly 70% of automotive organizations are already using or testing agentic AI in product development. The appeal is obvious, 83% of executives expect the technology to finally free their engineers from administrative gridlock so they can get back to core engineering.
For a select group of early adopters, that promise is already a reality. These companies are seeing 17% faster development cycles, a 19% boost in engineering productivity, and a 19% drop in time-to-market. But their edge doesn’t come from simply buying the latest AI tools. It comes from doing the unglamorous work, building rock-solid data foundations and enforcing strict governance.
For the rest of the industry, scaling AI is proving incredibly difficult, and the biggest roadblock isn’t the technology—it’s trust.
While engineers are perfectly happy leaning on AI for early-stage research and concept work, confidence plummets to just 40% when it comes to core development. When safety, quality, and real-world performance are on the line, you need absolute traceability. Right now, a staggering 95% of organizations admit their AI governance is still immature.
Beyond AI, automakers are also fumbling the basics of the software-defined vehicle loop. While connected cars are generating mountains of real-world data, 64% of companies admit they only use these insights for future vehicle programs. Instead of continuously improving the cars currently on the road, over-the-air updates are largely relegated to basic infotainment and security patches.
The ‘Rewiring automotive product development: how organizations can pull ahead in the AI race’, research mentions that the automotive industry’s biggest hurdles are now organizational, not technical. As Capgemini’s Laurence Noël points out, the companies that win the next decade won’t be the ones that adopt AI the fastest; they will be the ones that completely rewire their engineering workflows, break down data silos, and finally align their massive supplier ecosystems.
The research also points to the growing importance of ecosystem collaboration. As software-defined vehicles become more complex and development becomes increasingly distributed across OEMs, suppliers and technology partners, success will depend on greater alignment across the value chain.
However, only 30% of organizations report co-development models built around shared outcome metrics, while just 10% have established shared lifecycle accountability across their ecosystem. This highlights the challenge many organizations face in aligning partners around common goals and shared responsibility for product outcomes.
Methodology of the report
The Capgemini Research Institute surveyed 800 senior executives from the automotive industry, comprising 400 automotive OEMs, 160 tier-1 suppliers, and 240 technology providers.
Respondents represented organizations across North America, Europe, and Asia-Pacific, and included leaders from product development, engineering, technology, data and AI strategy, and related functions.
