A modern vehicle now runs on around 650 million lines of code, and software is becoming the part of the car that defines its value. For automotive OEMs and suppliers, the 2026 question is no longer whether to build software capability but how to build, integrate, and validate it without the cost overruns and recalls that software defects already cause. This is a data view of where the spend is going and what is actually being built. The numbers that frame the shift According to Statista, the average vehicle is expected to run on roughly 650 million lines of code in 2025, after the per-vehicle figure roughly doubled from about 100 million to 200 million between 2015 and 2020. The McKinsey Center for Future Mobility projects the automotive software and electronics market reaching around 519 billion US dollars by 2035 at roughly 4.5 percent CAGR, with software development (including integration, verification, and validation) carrying a revenue potential near 83 billion US dollars by 2030. Forecasts for the broader software-defined vehicle market diverge widely by analyst, which is itself a useful signal about uncertainty: estimates range from roughly 315 billion US dollars in 2025 (PS Market Research) to projections of 1.2 to 1.6 trillion US dollars by 2030 (MarketsandMarkets, BCC Research). The direction is consistent even where the magnitude is not: software is moving from a vehicle component to the vehicle’s defining system. Why downtime data belongs in this conversation Automotive software is built where automotive plants run, and the cost of getting it wrong is set by those plants. Siemens’ 2022 True Cost of Downtime study found the cost of an hour of downtime in automotive now tops 2 million US dollars, among the highest of any industry. Software that touches the line (manufacturing execution, supplier integration, and traceability) is