As AI Moves from Training to Inference, Optics Moves Closer to the Chip
Imec researchers argue that co-packaged optics will not be enough for future AI systems, pushing the industry toward 2.5D and eventually 3D optical I/O. The post As AI Moves from Training to Inference, Optics Moves Closer to the Chip appeared first on EE Times .
The increasing computational demands of AI inference, particularly with large models, are pushing current interconnect technologies to their limits, necessitating innovation in optical integration.
This development indicates a fundamental shift in how AI hardware will be designed, moving optics closer to silicon to overcome bandwidth and power constraints, impacting the future performance ceilings of AI.
The conventional approach of co-packaged optics is being superseded by more integrated 2.5D and 3D optical I/O solutions, fundamentally altering chip architecture for AI systems.
- · Optics manufacturers
- · Advanced packaging companies
- · AI chip designers
- · Hyperscale data centers
- · Traditional electrical interconnect providers
- · Older AI hardware architectures
Increased performance and energy efficiency for AI inference workloads.
New supply chains and manufacturing processes will emerge for integrated optical and silicon components.
This could enable the next generation of AI models that are orders of magnitude larger and more capable, further accelerating AI development.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at EE Times