Can AI functions in cars work without an internet connection?

Question

Accepted Answer

AI Functions in Vehicles Can Operate Independently of an Internet Connection

Artificial intelligence functions within vehicles are now capable of operating without a constant internet connection, a development driven by the integration of small language models (SLMs) directly into the vehicle's onboard systems. This advancement allows for core AI functionalities to remain accessible even in areas with limited or no network coverage, enhancing both user experience and system reliability. The technology leverages specialized AI models and powerful computing hardware designed for automotive applications.
Onboard AI Processing Enhances Reliability

The integration of small language models (SLMs), such as Microsoft's Phi-4-mini-instruct, into vehicles enables inference to occur directly within the car. This means that critical AI functions, including the configuration of vehicle settings or individual profiles via voice command, are not dependent on external cloud servers. Consequently, these central functions remain available and operational even when the vehicle experiences a limited or complete loss of internet connectivity, ensuring a consistent user experience.
Collaboration Drives In-Vehicle Generative AI

FEV is working with Microsoft and NVIDIA to embed Generative AI capabilities into vehicles. This collaboration focuses on utilizing NVIDIA GPU-accelerated compute power and AI model microservices to facilitate multimodal interactions, encompassing voice, text, and gesture. The goal is to create AI systems that are not only powerful but also efficient and responsive within the automotive environment.
SLMs Offer a More Efficient AI Approach

The use of small language models (SLMs) in vehicles increases the overall functional intelligence and improves the responsiveness and availability of AI-powered features. By processing AI tasks locally, this approach supplements or can partially replace cloud-based large language models (LLMs), depending on the specific use case. This strategy is designed to reduce backend and infrastructure expenses for original equipment manufacturers (OEMs).
Cost-Effective Scaling for Software-Defined Vehicles

This technological integration aids original equipment manufacturers (OEMs) in economically scaling the software-defined functions of their vehicles. By employing efficient, in-vehicle AI models that can operate independently of constant connectivity, automotive companies can manage costs more effectively while delivering advanced AI features to consumers. This approach ensures that the intelligence and capabilities of modern vehicles can be deployed and maintained more sustainably.

Reading history

Hide →

This history is cleared when the page is closed.

Can AI functions in cars work without an internet connection?

Loading

Was this article helpful?

Helpful

Not helpful

Report an issue

2

0

0

Share with friends