AI is poised to transform chip and semiconductor innovation across the full stack, from circuit discovery and architecture exploration to manufacturing, verification, and system-level optimization. As AI growth becomes increasingly constrained by hardware capacity and design complexity, accelerating semiconductor innovation has become central to the future of AI itself.
This full-day workshop, hosted as part of the AIAS+ 2026 Symposium — the Chen Institute Symposium for AI Advancing Science & Society (November 5–7, 2026, San Francisco, CA) — brings together researchers and practitioners advancing AI-driven chip design and semiconductor discovery. The workshop's primary focus is AI-native methodologies for electronic circuit and semiconductor system design, spanning physics-informed machine learning, graph neural networks, large-scale circuit datasets, and AI-assisted verification.
By uniting expertise across machine learning, circuit design, computer architecture, manufacturing, and scientific discovery, the workshop aims to catalyze a research community around AI-driven electronic and semiconductor systems, aligned with the AIAS+ 2026 vision of AI as a transformative partner in discovery.
🔬 AI for Analog/RF/Mixed-Signal Design
Physics-informed machine learning and graph neural networks for analog and RF circuit synthesis, topology selection, and performance prediction.
🧠 Physics-Informed & Knowledge-Driven Methods
Intelligent knowledge retrieval for design reuse, physics-aware models, and AI-assisted verification across the semiconductor stack.
📊 Datasets, Benchmarks & Foundation Models
Large-scale circuit datasets, standardized benchmarks, and foundation models enabling reproducible, scalable AI for hardware design.
🏭 Trustworthy & Scalable AI for Semiconductors
Broader AI methods for semiconductor materials, devices, manufacturing, packaging, and systems — with a focus on reliability and scalability.
