What is a software-defined vehicle?

Software-defined vehicles are the next evolution of the automotive industry. Traditionally, vehicle functions were tied to physical components and embedded systems with limited flexibility. SDVs instead rely on centralized computing platforms and modular software architectures. These systems allow over-the-air (OTA) updates, where automakers can deliver new features, updates and performance and safety enhancements through software, often remotely.

This modernization allows SDVs to evolve after purchase, much like smartphones. A vehicle might gain better navigation, improved energy efficiency or even enhanced driving modes through vehicle software updates alone, without having to visit a dealership. These capabilities also allow drivers to personalize their vehicles and subscribe to features on-demand, from advanced driver assistance systems to in-car entertainment upgrades.

IBM research predicts that 90% of all vehicle-related innovations are expected to consist of software in 2030.¹ And 75% of auto industry executives anticipate that the software-defined experience is going to be the core of brand value by 2035.²

A key part of this transformation is the reduction or elimination of many independent electronic control units (ECUs). ECUs are small computers that traditionally controlled individual vehicle functions such as braking, engine timing or climate control. For decades, automakers have added more ECUs to support new features. Some vehicles had more than 100 of these units.

Today, many are replaced with fewer, more powerful central computers that manage multiple systems at once. This reduces complexity and allows vehicle systems to work together more smoothly. It also supports innovations like autonomous driving, predictive maintenance and real-time data integration with cloud services.

SDVs, connected vehicles and autonomous vehicles are closely related but not the same.

Connected vehicles are cars equipped with internet access and vehicle to everything (V2X) communication. V2X allows them to share data with other vehicles, road infrastructure and external systems (for example, a toll payment system or mobile app) and the cloud. Over 327 million connected vehicles are expected to be in service by 2027.³

Their connectivity can help to reduce accidents and improve traffic flow. Both SDVs and connected vehicles rely on software-driven functions, real-time data and cloud integration.

Most modern SDVs also use V2X, so the difference between them and connected vehicles is small. Connected vehicles prioritize external communication, while SDVs rely on internal software architecture that upgrades core functions through OTA updates. In other words, all SDVs are connected, but not all connected vehicles are SDVs.

Autonomous vehicles use sensors, cameras and advanced software to detect their surroundings and drive themselves without human input. This ability is only possible within an SDV framework, which uses centralized computing to manage vehicle systems. So, while not all SDVs are autonomous, all autonomous vehicles are SDVs, as SDVs provide the software foundation needed for autonomy.

SDVs also complement the rise of electric vehicles because both emphasize efficiency, connectivity and reduced environmental impact.

The software-defined vehicle (SDV) model marks a major shift in how cars are designed, built and experienced. Automakers now operate more like software companies, with vehicles functioning as a dynamic, updatable platform. Tesla pioneered this model, demonstrating how software-driven upgrades can unlock new revenue streams and build brand loyalty.

OEMs (original equipment manufacturers) are shifting their focus from mechanical engineering to digital innovation. This new approach reshapes expectations and unlocks business models based on software, not just hardware.

SDVs move the core value of a vehicle from its mechanical parts to software that can be improved over time. Features, performance and even compliance with new automotive regulations can be added or upgraded remotely, without changing physical components. This ability can extend the useful life of a vehicle and keep it current longer.

SDVs also play a central role in advancing safety, automation and connectivity. Their software-based vehicle architecture enables features like advanced driver assistance systems (ADAS), self-driving capabilities and V2X communication.

The SDV development process is faster and more flexible, too. Engineers can use virtualization (technology that enables the creation of virtual environments) and simulation to test software in digital environments before any hardware is built. This cuts time, cost and risk, similar to how software development works in the tech industry.

Ultimately, SDVs are not just modern cars. They are intelligent platforms that can grow and adapt, enabling safer, smarter and more sustainable transportation.

The key characteristics of SDVs redefine what vehicles can do and how they are designed, operated and monetized. They include:

SDVs consolidate vehicle functions into powerful central or zonal computers, replacing dozens of distributed ECUs. This architecture allows for more efficient data processing and coordination across systems.

Software can be updated remotely to improve vehicle performance, fix bugs, add features or enhance security without requiring a service visit.

SDVs use modular software platforms that are decoupled from hardware, enabling easier upgrades, longer vehicle lifespans and flexible deployment of features.

SDVs use a stack that typically includes an embedded operating system (like QNX or Linux), middleware, application frameworks and user-facing apps. These systems and tools are all designed to be updatable.

SDVs use virtualization to isolate critical functions (like safety features) from noncritical ones (like infotainment systems). This separation improves security and helps ensure that problems in one system don’t affect the others.

SDVs are built to communicate in real time with cloud-based services, vehicle to infrastructure (V2I), vehicle to other vehicles (V2V) and mobile devices. This connectivity enables services like live navigation, remote diagnostics and intelligent routing.

79% of automotive OEM executives expect their SDV efforts to progress in the next three years. 76% believe artificial intelligence (AI) s set to contribute to this progress.⁴ Machine learning and AI are integrated for real-time sensor fusion (combining data from multiple sensors to make fast, accurate decisions). AI also supports predictive maintenance, personalization and autonomous driving functions—use cases that support its growing role in modern vehicle platforms.

Many SDVs allow users to purchase or subscribe to features post sale, such as advanced cruise control, heated seats or performance modes. Today, digital- and software-related revenue represents 15% of total auto industry revenue. This share is expected to rise dramatically to 51% by 2035.⁴

With their centralized architecture and software-based control, SDVs are better suited to support ADAS, self-driving capabilities and evolving safety standards.

Given their connectivity, SDVs are built with embedded security features to protect against threats. These features include secure boot (where the server boots only trusted software), encrypted communication, real-time monitoring and intrusion detection systems. 86% of auto industry executives agree that security, assurance and trust are brand attributes that differentiate their organizations.³

Automakers can extend the useful life of a vehicle by continuously evolving it through software, helping reduce waste and support sustainability goals.

SDV platforms use virtualization, simulation tools and generative AI in automotive to explore design alternatives, simulate edge cases and support system validation before physical prototypes are built.

Modern cars are no longer just machines, they're rolling computers. But how those computers are organized has changed dramatically.

In traditional vehicles, every major function had its own tiny computer, an ECU. One ECU might handle your brakes, another the airbags, another the radio and so on. Some cars had 100 or more of these ECUs. While that setup allowed automakers to add features over time, it also made vehicles complex and heavy, with miles of wiring running between all those devices.

To reduce the complexity, the industry introduced domain controllers. These controllers are like middle managers that organize related ECUs by area of responsibility—so one domain controller might manage all things related to driving assistance, while another handles infotainment. This approach reduced the number of ECUs but didn’t fully solve the complexity problem.

The next significant leap is happening now: automakers are moving to high-performance computers (HPCs) and zonal architectures. Instead of dozens of scattered ECUs, a few powerful central computers (HPCs) are used to run many functions at once, like a car's main brain. Supporting the HPCs are zone controllers placed in different physical areas of the vehicle, which manage local sensors and devices, then pass information to the central HPC. These local systems often include radar, camera and LIDAR sensors, which feed detailed environmental data to the vehicle’s central computing unit.

This approach reduces wiring (which cuts cost and weight) and makes the car’s system easier to manage. It also opens the door to OTA updates, so your car can get new features or fixes without going to the dealership. This new architecture supports future technologies like self-driving systems, which require centralized, high-speed computing. It also enables vehicles to more fully participate in the Internet of Things (IoT), exchanging data with connected devices, infrastructure and services in their environment. All these capabilities are made possible by AI and recent advancements in automotive processing.¹

Just like the hardware in cars has evolved, so has the software. In traditional vehicles, the software that controlled each ECU was tightly bound to that specific piece of hardware. Changing or updating it was difficult and time consuming and often required physical access to the vehicle.

To help manage this shift, the industry introduced a standard called AUTOSAR (AUTomotive Open System ARchitecture). It was designed to make automotive software more reusable and consistent across different brands and suppliers. AUTOSAR worked well for traditional functions like engine control or airbag systems, where stability and safety are critical and changes are rare.

But modern vehicles demand more flexibility. Features like advanced driver assistance, in-vehicle voice assistants and cloud connectivity need frequent updates and more complex software, like what’s on smartphones or servers. So the newer AUTOSAR Adaptive was designed to work on high-powered computing platforms, built on familiar technologies like Linux and Ethernet to enable more dynamic, real-time services. It also supports cloud-native approaches, where the software is designed to run easily across connected systems and can be updated or scaled more efficiently.

Looking even further ahead, carmakers are starting to adopt techniques from the tech industry such as containerization. Containers are lightweight packages of software that include everything needed to run an app, making them easy to test, update and deploy. They’re like smartphone apps—modular, isolated and updateable without affecting the rest of the system. Combined with strong APIs (interfaces that let software components talk to each other), this approach allows vehicles to shift from rigid, monolithic codebases to flexible, microservices-based systems.

Software in cars is evolving from something static and slow to change into something agile, smart and always improving. It’s the app-store experience for your dashboard.¹

SDVs offer a range of benefits that enhance safety and performance and transform the overall driving experience.

Continuous connectivity: Always-on connections keep vehicles in touch with cloud services, navigation updates and traffic data. This connectivity enhances the driving experience and enables real-time responsiveness.

Enhanced performance and efficiency: Smart software can fine-tune driving dynamics, battery usage and engine performance in real time. Depending on the settings, this adaptability can lead to better fuel economy or battery life or a more responsive driving experience.

Faster innovation and development: Automakers can design, test and roll out software faster through virtualization and modular development. This acceleration shortens the time from idea to real-world feature.

Improved safety: SDVs power advanced safety systems like emergency braking, lane-keeping and collision avoidance. These features rely on real-time data and fast decision-making, making roads safer for everyone.

New revenue opportunities: Manufacturers can generate ongoing income by offering subscriptions, on-demand upgrades or app-based services. These offerings turn vehicles into long-term platforms, not just one-time sales and are popular with automakers but not always with car owners.

Predictive maintenance: SDVs can monitor their own systems and detect issues before they become serious. This helps reduce breakdowns, avoid expensive repairs and keep the car running smoothly.

Personalized user experience: Drivers can tailor their vehicle’s settings to their preferences, for example the dashboard layout or their choice of in-car entertainment. The car can also remember different profiles for different drivers.

Remote feature updates: Just like smartphones, SDVs can receive software updates over the air. This means new functions, updates and improvements can be delivered long after the car leaves the factory.

SDVs bring numerous benefits, but they also present significant challenges. One of the biggest hurdles is the shift from traditional mechanical systems to digital architectures. In fact, 79% of executives cite the technical complexity of separating hardware and software layers as challenging.² Other, more specific drawbacks don’t negate the promise of SDVs but highlight the need for careful design and robust governance as the industry evolves. These challenges include:

Consumer pushback on monetization models: Subscription-based access to features that were once standard (for example, heated seats and adaptive cruise control) might frustrate customers and hurt brand perception.

Cybersecurity risks: With increased connectivity, comes greater vulnerability. SDVs are exposed to potential cyberattacks targeting vehicle controls, data privacy or cloud-based services. They require constant vigilance and advanced security frameworks.

Data privacy and ownership issues: With SDVs constantly collecting data, concerns about how that data is stored, used and shared—particularly without explicit consent—pose ethical and regulatory questions.

High development and maintenance costs: Developing, testing and validating SDV platforms is expensive and time-consuming. Safety-critical functions and over-the-air infrastructure updates are especially complex.

Increased software complexity: SDVs shift the burden from mechanical to software complexity. Managing millions of lines of code across multiple systems, layers and vendors creates integration challenges and increases the potential for bugs or failures.

Talent shortages: The auto industry now requires software engineers, AI specialists and cybersecurity professionals—talent more traditionally found in tech companies. Many automakers are still building this internal capacity. 74% of executives say that their mechanical-driven culture is strong and difficult to change. They need employees skilled in both software development and traditional vehicle engineering, but don’t expect to build the workforce required to achieve their software-defined product goals until 2034.²

Regulatory and legal hurdles: Updating vehicle behavior through software introduces new legal and regulatory questions. It especially raises issues regarding accident liability, data ownership and compliance with evolving safety standards.

Reliability concerns with AI and automation: As SDVs incorporate AI-based decision-making (for example, for ADAS or autonomous driving), questions remain around explainability, predictability and how to manage issues like system overrides.

System compatibility fragmentation: The lack of standardization across platforms, operating systems and cloud environments within the broader ecosystem can make it difficult to support compatibility and scalability across different vehicle models and regions.

Update management risks: While over-the-air updates are convenient, poorly managed updates might lead to system failures and user frustration.

The future of SDVs is one where the car becomes a connected, intelligent platform—not just a machine. Software shapes the driving experience more than hardware. Cars are updated, personalized and improved through software, much like a smartphone. Buying a car might feel more like subscribing to a service, with new features and upgrades delivered over time through OTA updates.

As this shift continues, the auto and tech industries are set overlap even more. Technologies like cloud computing, AI, 5G and edge computing are poised to power SDVs. Automakers are also expected to rely on a hybrid cloud approach that uses a mix of public and private cloud systems to manage data, support updates and deliver new services. To keep up, OEMs must operate more like tech companies by adopting faster development cycles, strong cybersecurity and flexible, modular systems.

This evolution is expected to transform the customer experience. Drivers are set to expect regular updates, remote support and personalized features. Real-time data is anticipated to enable predictive maintenance, smarter navigation and custom settings. As self-driving technology advances, SDVs are likely to emerge as the platform for introducing and fine-tuning those capabilities—particularly in urban areas and shared mobility fleets.

SDVs are a key part of smart mobility. They connect with traffic systems, energy grids and digital services to support safer, more efficient and more sustainable transportation. This shift is expected to transform not just how we drive, but how we move, own and interact with vehicles.