The AI4SDV research project has launched to build an open, AI-assisted engineering toolchain for software-defined vehicles (SDVs), bringing together more than 50 partners from industry and academia across Europe. The initiative, funded under the Chips Joint Undertaking as part of Horizon Europe, aims to accelerate vehicle‑software development while keeping safety, traceability and verification processes up to date.
Project scope, partners and targets
AI4SDV started in September 2026 and will run for 36 months. The consortium says the programme’s total budget exceeds 550 million and will contribute roughly 350 person-years of work to build an open and extensible engineering platform. The project lists major automotive and supplier partners including Mercedes‑Benz Tech Innovation, Renault/Ampere, Infineon Technologies, dSPACE, Scania/Traton, TTTech and Thales alongside research organisations such as the Technical University of Munich and Fraunhofer IESE. The High‑Performance Computing Center Stuttgart (HLRS) will provide compute capacity for training and operating large AI models used by the project.
AI4SDV aims to apply AI across the full engineering cycle for selected SDV functions: automated code and test generation, architecture error detection and automated support for safety and change‑impact analyses. The consortium sets measurable internal targets: AI could support up to 80% of engineering tasks in the long term and the number of errors in first complete functional versions should fall by more than 30%.
Safety, traceability and the human role
Fraunhofer IESE, which published the project announcement on 28 October 2026, emphasises that safety engineering, verification and compliance must keep pace with faster AI‑assisted development. Jan Reich, head of Safety Engineering at Fraunhofer IESE, is quoted on the need to make AI outputs understandable and reviewable by engineers, saying safety cases must not “lag behind by months” if development accelerates.
The project therefore focuses on methods that extract safety‑relevant knowledge from standards and engineering artefacts, link safety cases across diverse artefacts, and enable engineers to review and correct AI‑generated results. The stated objective is to reduce the effort for safety‑related engineering tasks while increasing quality and traceability of outcomes.
What this means for the market
AI4SDV tackles a central bottleneck for European vehicle makers and suppliers: the rising cost and complexity of software development for SDVs and advanced driver assistance systems. By packaging AI‑assisted tools into an open engineering stack and providing European compute infrastructure, the project aims to help OEMs and tier‑one suppliers shorten development cycles and scale software‑driven features without weakening safety evidence. For Central and Eastern European suppliers and plants that already produce software, electronics and ADAS components for global OEMs, an open European toolchain could lower integration costs and make it easier to participate in SDV development.
However, toolchain uptake will depend on how well AI4SDV can demonstrate traceable safety evidence that meets regulators’ and OEMs’ requirements. If successful, the platform could reduce time‑to‑market for features such as in‑car AI assistants, OTA updates and advanced ADAS functions — areas where European manufacturers and suppliers are already investing heavily.
Sources: Fraunhofer IESE, HLRS
Partner note: EastAutoNews technology partner DaVinci AI develops AI solutions for the entire automotive industry – from market and dealer analytics to autonomous AI agents and image and video production.
