3 AI Use Cases Advancing the Automotive Industry

According to some predictions about when fully autonomous vehicles would be available, we should all own at least one by now. That’s obviously not the case—yet. But automotive original equipment manufacturers (OEMs) have made tremendous progress, thanks to AI accelerating advances in capabilities, including design, user experience, navigation, safety, predictive maintenance, and driver wellness monitoring.

What’s exciting is that we’re still in the early stages of development. Today, most cars operate at Level 2 or Level 3 of SAE International’s Levels of Driving Automation[1] (autonomous emergency braking or accident-avoidance systems, respectively). But auto OEMs have developed Level 4 (driverless, but currently limited to robotaxis) cars and are developing Level 5 (fully autonomous) vehicles, which operate without control equipment such as steering, braking or acceleration. Cars at any of these levels may also be referred to as software-defined cars or connected vehicles, since they leverage software and add Over-The-Air (OTA) updates to control and enhance vehicle operations, instead of hardware like traditional vehicles use.

The higher the level of automation, the higher the degree of complexity and interoperability requirements the manufacturer needs to plan for. To meet these requirements, they rely on seamless, instant and adaptable connectivity across an ecosystem of industry partners and service providers, from established cloud and network providers to emerging AI specialists.

Together, these partners will fulfill the promise of autonomous driving, but they’ll need to be able to connect in a carrier and vendor-neutral environment to do it. Further, running connected vehicles safely and delivering optimal user experiences requires low latency and stable connectivity, which can be enabled with distributed AI infrastructure.

Enabling and innovating autonomous vehicle capabilities

Various types of AI technology can enhance the capabilities of autonomous vehicles. Machine learning has traditionally been the primary AI technology used by auto OEMs, but they’re increasingly incorporating GenAI and LLMs for natural language processing, data analysis and insights generation. Auto OEMs use GenAI to extract insights and identify patterns from sensors that collect various types of data, which enable real-time alerts to other drivers and further improve the safety and efficiency of their systems. For instance, if a car detects a road hazard, it will send the details to the closest edge data center, which will initiate a prompt to distribute a warning to other vehicles in the area.

Auto OEMs use this newly generated data to fine-tune AI models, then deploy them back to edge data centers to replace older models currently used for AI inference. To support different training and inference workloads, and thereby improve convenience and safety, OEMs need to move datasets quickly, securely and intelligently across clouds, centralized infrastructure and edge locations.

Since vehicles travel between cities or countries, it’s also necessary to ensure the connectivity switches from one edge data center to another. Interestingly, a car’s data may be collected and hosted in more than one data center on any given journey. This ensures that data continues to flow to and from those vehicles for consistent, seamless operations. Since the cars may be running on mobile networks that belong to different telcos or mobile network operators, it’s also necessary to interconnect those networks.

Industry partners and providers are collaborating with members of organizations, such as the Automotive Edge Computing Consortium[2], to influence the evolution of edge network architectures and computing infrastructure, enabling high-volume data services to deliver smarter, more efficient connected vehicle services. Equinix is a member of this organization and has helped develop proofs of concept that demonstrate commercial use cases, including deployment requirements.

Auto companies may not have sufficient data to train their AI models or may lack the necessary AI models. To overcome these limitations, they can participate in various data and AI model marketplaces on vendor-neutral platforms, where they can explore options for exchanging data and AI models that will assist in the ongoing training, adaptation and adoption of automated vehicles.

Three use cases for AI in the automotive industry

Auto OEMs are increasingly relying on AI to accelerate the development of new capabilities for their vehicles, with a focus on enhancing user experience and safety. While they’ve achieved meaningful milestones, they continue to focus on realizing evolving opportunities and raising the bar on what’s possible. Let’s explore three use cases where AI is advancing progress in the automotive industry.

Enhancing the user experience

Global auto enterprises like Hyundai Motor Group are continuously focused on improving customer experience through personalized in-car services, including entertainment, navigation, driver wellness monitoring, and context-aware recommendations based on driver profiles and preferences. Hyundai’s proprietary HCloud platform powers advanced connected car services (CCS) for over 10 million subscribers, enabling seamless connectivity for entertainment and mobile applications.

HCloud was developed in response to the growing demand for real-time data processing, seamless connectivity and scalable infrastructure, driven by rapid advancements in connected and autonomous vehicles. The company leverages Equinix IBX® global data centers and Equinix Fabric® virtual networking to create a hybrid multicloud architecture that connects HCloud to multiple cloud providers, including AWS. This infrastructure is accelerating the global rollout of Hyundai’s connected car services with reduced latency, reliable connectivity and consistent service coverage worldwide. Read the press release.

Improving safety through AI evaluation

Tensor has announced the launch of the world’s first fully autonomous personal Robocar in 2026. Delivering real-time Level 4 autonomy in this groundbreaking vehicle demands extremely efficient and rigorous safety evaluation at every stage of development. Their AI evaluation workloads are strategically hosted in Equinix AI-ready data centers. Specifically, Tensor’s Middle East deployment at Equinix in Dubai enables real-time data processing and model evaluation close to vehicle operations, thereby helping to ensure safer, more accurate and ultimately, more reliable driving for their customers.

Accelerating the ADAS design process

Continental’s ADAS team needed to process over 150 terabytes of image and sensor data, collected from vehicles worldwide and stored in a centralized repository, to inform design decisions made by hundreds of engineers globally that would enhance the safety of connected and autonomous vehicles. The company leveraged Equinix’s distributed AI infrastructure to build and interconnect an AI-driven NVIDIA DGX GPU cluster with IBM Elastic Storage System 3000. This scalable, future-proof AI approach reduced training time from weeks to days while providing real-time data access, enhanced security and improved data privacy controls to accelerate safety standard development. For more details, read the case study.

Get connected with distributed AI infrastructure

Auto OEMs need distributed AI infrastructure wherever their data lives, whether it’s for secure decision-making, AI training or inference, vehicle safety or a seamless user experience. Equinix Distributed AI^™ is a comprehensive solution that helps them deploy infrastructure to support connected vehicles with low-latency connectivity and continuous development of capabilities.

Equinix provides the global reach necessary to support distributed AI, with 270+ AI-ready data centers across 77 strategic markets worldwide. The Equinix ecosystem includes more than 10,000 enterprises and service providers, ranging from established cloud and network service providers to emerging AI specialists. This makes it easy for our customers to find the right partners, since those partners are likely already deployed at Equinix. Auto OEMs can build a foundation on neutral infrastructure that helps them meet evolving requirements without being constrained by vendor limitations or inflexible technology choices.

Learn why the future of AI infrastructure is distributed: View videos with Equinix executives for an introduction to our Distributed AI solution.

You may also be interested to read our Distributed AI solution brief.

[1] SAE Levels of Driving Automation™ Refined for Clarity and International Audience, SAE International, May 3, 2021.

[2] Automotive Edge Computing Consortium.