In the AI era, your automotive cloud strategy must be engineered for execution. You can’t win with AI if your cloud isn’t up to speed.
In NTT DATA’s global research into the increasingly intertwined relationship between cloud technologies and AI, respondents see success with AI as being predicated on success with cloud.
But our report, Cloud-led innovation in the era of AI: The new rules for driving value with cloud, also reveals a clear gap between cloud ambition and cloud readiness: Only 14% of surveyed organizations (and only 10% in manufacturing and automotive) report that they are “cloud-evolved” — the most advanced when it comes to cloud adoption and impact — while 99% say AI, including agentic AI, has increased their need for cloud investment.
How the automotive industry is adapting to AI
For automotive manufacturers and suppliers, that gap is practical. The industry is moving from hardware-centric vehicle programs to software-defined, electrified and connected mobility ecosystems. AI, cloud and software-defined mobility are no longer separate conversations. Vehicles are becoming more connected, more data-intensive and more dependent on software, making cloud part of the operating architecture of automotive enterprises.
That matters because for AI to create value most effectively, it must work across engineering, production, supply chain, vehicle platforms and customer experience. Cloud provides the governed execution environment that makes such work possible, connecting data, applications, controls and computing so that AI can support decisions throughout the automotive operating model.
As cloud becomes increasingly foundational, automotive cloud strategy now has to support this shift to enable safety, security, compliance, latency, data ownership and continuous software delivery.
Our report identifies six rules for creating cloud value in the AI era. Here, we’ll focus on those that determine whether AI can operate safely and effectively in a software-defined mobility ecosystem — rules relating to architecture, platform-led operations and security.
Platform-led cloud turns automotive complexity into execution
Automotive companies have abundant data from multiple sources, but that wealth of data comes at a cost. Datasets from engineering, manufacturing, supply chain, vehicle platforms, dealer networks and customer channels often involve different rhythms, architectures and governance models.
This fragmentation becomes more consequential as vehicles become software defined. What used to be primarily a software update now requires a massive, coordinated effort — including engineering validation, release governance, cybersecurity controls, vehicle connectivity, customer experience and even supplier coordination.
The same is true for AI-enabled quality inspection, predictive maintenance, connected services and increasingly autonomous workflows across the enterprise.
This makes a platform-led approach essential. Over the next two years, manufacturing and automotive organizations expect the use of fully managed and automated cloud platforms delivered by strategic partners to more than double, from 11% to 27%. It reflects a practical reality: these organizations need a consistent operating layer that connects applications, data, infrastructure, networks and AI-driven actions throughout the business.
As AI workloads scale across engineering, production and connected services, costs can quickly spiral across the business. Almost 6 in 10 automotive and manufacturing organizations (58%) say cloud cost management is a challenge and that they should apply more focus to cloud cost optimization.
For the automotive industry, where reliable repeatability is key, a platform-led model enables teams to innovate locally in the factory, in engineering and in connected services while still operating within a common framework for security, reliability, cost control and governance.
Cloud architecture determines what AI can safely do
Cloud architecture decisions can make or break your business. When determining where to place AI workloads, 52% of automotive and manufacturing organizations cite data security, privacy and compliance as a leading consideration. Another 53% point to the need for greater control over infrastructure and customization.
The real question is not simply where a workload can run, but where it should run based on performance, control, risk, integration and lifecycle needs. An AI system that accelerates requirements management in R&D has different constraints than an AI-enabled visual quality inspection system. A connected vehicle platform has different requirements than a customer personalization engine. Software-defined vehicle capabilities, including over-the-air updates and adaptive in-vehicle experiences, depend both on cloud connectivity and on disciplined controls regarding release processes, data integrity, safety and operational continuity.
In this industry, architecture is where the operating model starts to change — and when cloud decisions are fragmented, AI stays trapped in isolated use cases. But when the architecture is designed around the full vehicle, factory and customer ecosystem, AI can help coordinate decisions across domains to enable shorter development cycles, improve production responsiveness and make connected experiences more adaptive over time.
Security makes scale possible
All industries rank security first among their priorities for cloud. Security, governance, risk and compliance concerns regarding autonomous agents are also the top challenges in adopting agentic AI in cloud-based solutions over the next 12 to 18 months.
For automotive companies, this priority spans the full ecosystem: enterprise IT, engineering environments, manufacturing systems, connected vehicles, software platforms, suppliers, infrastructure partners and customer-facing services. A weakness in one area can affect intellectual property, production continuity, regulatory compliance, customer trust or vehicle-related systems.
As AI agents begin to influence decisions, the need for cybersecurity increases. In the automotive industry, the issue is not only whether an agent can access data. It’s also what the agent can decide, what action it can trigger, which system it can affect and when human oversight is required. That distinction matters when software updates, test conditions, production processes or customer-facing services are involved.
Security, in this sense, is not a brake on innovation. It is what gives automotive organizations permission to move faster with confidence.
How cloud decisions drive automotive execution
Success in the next phase of automotive cloud strategy will be measured by whether cloud can help connect vehicles, factories, software teams, partners and customers into a more intelligent operating model.
AI can change how vehicles are engineered, produced, updated and experienced, but it will not deliver that value through isolated pilot projects or disconnected automation. It requires a cloud foundation designed for integration, governance and execution.
WHAT TO DO NEXT
Access our report, Cloud-led innovation in the era of AI: The new rules for driving value with cloud, for a closer look at our research.