The automotive industry is undergoing a seismic shift as Large Language Models (LLMs), Vision Language Models (VLMs), and multimodal AI applications move from research labs into vehicles, factories, and customer experiences. These technologies promise transformative capabilities—from predictive maintenance and autonomous driving to intelligent voice assistants and real-time visual inspection—but they also introduce unprecedented strain on memory subsystems. This talk explores how the explosion of multimodal workloads are reshaping memory architectures across the automotive value chain. We’ll examine the specific challenges Tier 1 suppliers and OEMs face in deploying these models at scale, including bandwidth bottlenecks, latency constraints, and thermal limits. We’ll also highlight how the market is responding—with innovations in memory hierarchy and system-level optimization—to meet the demands of AI-driven automotive platforms.

• Large Language Models (LLMs), Vision Language Models (VLMs), and multimodal AI applications move from research labs into vehicles, factories, and customer experiences
• Innovations in memory hierarchy and system-level optimization – to meet the demands of AI
• Explore how the explosion of multimodal workloads are reshaping memory architectures across the automotive value chain