The shift reflects a broader transformation underway in the industry. As automakers move toward so-called software-defined vehicles — in which features and performance are governed as much by code as by mechanical engineering — cars are increasingly equipped with sensors, connectivity tools and onboard computing systems. These vehicles continuously produce data, from driving behavior and system diagnostics to environmental conditions, much of it used to refine performance, enable new services and train artificial intelligence models.
But that surge in information is straining existing infrastructure. Traditional internet-of-things frameworks, the report argues, were not designed to handle the scale or cost demands of automotive data, which can reach tens of gigabytes per vehicle each day.
The consortium’s proposal centers on distributing the burden of processing and transmitting that data across multiple layers, rather than relying primarily on centralized cloud systems. In this model, vehicles would communicate directly with one another in some cases, exchanging data locally. Nearby edge computing resources — such as localized data centers or wireless access points — would handle additional processing, while cloud networks would provide coordination and long-term storage.
Advocates of the approach say it could reduce network congestion and energy use while improving the speed at which data can be analyzed and acted upon. It also reflects a growing consensus that future automotive services — from advanced driver assistance to personalized in-car experiences — will depend on faster, more efficient ways to move and process data.
“Access to high-quality, large-scale data will be essential,” said Dr. Ryokichi Onishi, the consortium’s board chair, in a statement accompanying the release.
The proposal arrives as automakers, technology firms and telecommunications providers race to define the infrastructure behind the next generation of connected vehicles. Whether a distributed, data-first model becomes the industry standard remains to be seen. But the direction is clear: as cars become more intelligent, the systems that support them must evolve just as quickly.