Modern ADAS/AD platforms can involve 10,000+ customer requirements, expanding into tens of thousands of derived requirements, architectural elements, interfaces, and traceability relationships. As requirements and architecture evolve concurrently, maintaining consistency manually becomes increasingly difficult, driving rework, cost, and program risk. This presentation presents a practical AI-assisted systems engineering approach to continuously identify requirements-to-architecture gaps within Automotive SPICE workflows. It demonstrates how AI can augment engineering review, strengthen traceability, and detect inconsistencies earlier while preserving human oversight for safety-critical decisions. Attendees will gain a scalable framework for managing growing intelligent-vehicle complexity while maintaining engineering rigor.

• Understand why requirements-to-architecture gaps grow rapidly as ADAS/AD requirements and system complexity scale.
• Learn how AI-assisted analysis can continuously identify missing, inconsistent, or weak requirements-to-architecture relationships.
• Apply AI-assisted systems engineering within Automotive SPICE workflows while preserving traceability and process compliance
• Identify where AI adds engineering value—and where human judgment remains essential for safety-critical ADAS/AD decisions.
• Learn a scalable approach to reduce manual review effort, late rework, development cost, and program timing risk.