Highlights from IEDM 2025: 3D NAND is suddenly relevant again, interconnect metals beyond copper are emerging, 2D materials that could replace silicon, and more
IEDM 2025 Round-Up — It's an odd time in the chipmaking industry. On one hand, we are ramping into the biggest supercycle ever seen.
Context & Ripple Effects
IEDM’s research agenda now spans memory, wiring and channel materials rather than transistor geometry alone. That extends the prior VLSI discussion of DRAM scaling beyond 1x nodes and Intel’s interconnect-scaling research.
The relevance is heightened by coverage identifying advanced logic wafer availability as an AI bottleneck, making progress across other parts of the semiconductor stack consequential rather than merely academic.
First-order effects
- 3D NAND, alternative interconnect metals and 2D channel materials receive renewed technical attention from device, process and materials teams; the report signals research momentum, not immediate production substitution.
- Interconnect scaling becomes a more explicit design constraint alongside transistor scaling, reinforcing the importance of work such as foundry research on new wiring approaches.
Second-order effects
- Equipment, materials and memory suppliers face pressure to support more heterogeneous process roadmaps if research programs move from demonstrations toward integration.
- Chip designers may place greater value on memory density and interconnect advances when leading-edge logic capacity remains constrained, as highlighted in coverage of N3 capacity limits.
Third-order effects
- If these research directions mature, semiconductor scaling will increasingly depend on coordinated advances in materials, memory architecture, interconnects and packaging rather than silicon transistor shrinks alone.
- The industry’s innovation pipeline may diversify beyond a small set of leading-edge logic nodes, though commercialization timelines and manufacturability remain uncertain.
The trend: This is one data point in the shift from transistor-centric scaling toward whole-stack materials and architecture innovation under AI-driven semiconductor demand.