January 19, 2026

What’s really holding back Software-Defined Vehicles

By Karim Jeribi, Europe General Manager, Consulting & Engineering Services at DXC Technology

The automotive industry has spent the last several years accelerating its shift toward software-defined vehicles (SDVs) – vehicles that can update over time, enhance their original features, and add new capabilities throughout their lifespan. But as automakers push to deliver vehicles that can evolve continuously, many are discovering that their biggest challenge isn’t inventing new capabilities. It’s integrating them.

Integration means ensuring that the many software systems, components, and features inside a vehicle work together seamlessly. Modern vehicles can have hundreds of electronic control units and software modules that must communicate reliably and predictably. Without the right architectural foundation, an automaker’s SDV program becomes slower, more expensive, and harder to scale. A change to one subsystem can unintentionally break another if they're not properly integrated, and security vulnerabilities can emerge when systems aren't designed to work together from the start.

To move forward, industry leaders must treat integration as the defining challenge of the SDV era. By rethinking how vehicles are architected, developed, and maintained throughout their lifecycle, they can innovate faster and build more resilient platforms. 

Why SDV integration has become so difficult

Today’s integration problem are decades old. Traditional vehicle development was built around hardware. Electronic control units (ECUs) were designed to perform specific functions like braking, steering, infotainment—often developed independently by different teams and suppliers. That model worked when software played a limited role. In the SDV era, it no longer does.

Modern vehicles now function more like rolling data centers — combining infotainment, advanced driver assistance systems, connectivity, AI services, and real-time control software. In vehicles with autonomous driving capabilities, these systems can generate terabytes of data, which must be processed, managed, and governed across different operating systems, hardware platforms, and development toolchains.

This architectural shift brings unprecedented complexity: Software and hardware timelines drift, teams work in silos, and integration and validation are often pushed late into the development cycle where problems are significantly more expensive and difficult to fix. Compliance and cybersecurity demands add further pressure, especially as software stacks become larger and more interconnected.

For many automakers, this fragmentation has become the hidden tax on SDV progress.

What makes SDV integration especially challenging is it’s no longer just an engineering concern; it’s a leadership imperative. Automakers are being asked to deliver vehicles that can be updated long after leaving the factory floor, support new functions, and maintain safety and regulatory compliance while still differentiating their brand experience. Meeting those expectations requires more than new tools. It demands new operating models, closer collaboration across software, hardware, and product teams, and a shift toward treating integration as a core capability designed into the vehicle architecture from the start, not patched together at the end.

Building the right foundation for SDVs

Addressing integration complexity starts with architecture. The industry is steadily moving away from ECU-based designs toward more centralized computing architectures. This approach allows multiple vehicle functions to run on shared hardware while remaining isolated from one another.

Virtualization plays a critical role here. By using virtual machines, automakers can ensure that non-critical systems, like infotainment, never interfere with safety-critical functions like braking or steering. This separation allows faster innovation without compromising reliability or compliance.

Equally important is abstraction. Middleware and hardware abstraction layers allow software to communicate across domains while remaining decoupled from specific chips, displays, or sensors. This flexibility reduces vendor lock-in and makes platforms more resilient to future hardware changes.

Together, these foundations turn integration into a manageable, scalable process.

Automation turns integration from a bottleneck into a strength

Even with the right architecture, manual integration cannot scale SDVs at speed. That’s why leading automakers are adopting software-factory approaches built on automation and continuous integration. Instead of waiting for physical hardware, teams can validate software early by using virtual targets, detect issues sooner, and reduce late-stage surprises.

Automated pipelines also improve software traceability, a critical requirement as safety, cybersecurity, and regulatory expectations increase. When updates can be tested, validated, and deployed continuously, vehicles are no longer frozen off the assembly line and can evolve throughout their lifecycle.

This approach doesn’t just improve efficiency; it fundamentally changes how quickly automakers can respond to customer expectations, technology advancements, and market shifts.