CASPA CEO Summit 2026
| title | author | length |
|---|---|---|
| Michael Hurlston, Lumentum Holdings, Inc. | CASPA CEO Summit 2026 | Michael Hurlston | 20m |
Optical interconnects are the asserted path to scale AI because copper hits a “copper wall” at multi‑terabit speeds, so fiber replaces electrical links across… | ||
| Irving Tan, CEO, Western Digital | CASPA CEO Summit 2026 | Irving Tan | 13m |
Forecasts put data generation at 3x to over 700 zettabytes in three years and storage needs rising to >26 ZB, with demand tracking >25% CAGR over five years… | ||
| Sunghyun Park, Rebellions | CASPA CEO Summit 2026 | Sunghyun Park | 18m |
Claims that AI inference economics hinge on token per dollar per watt and that infrastructure buildout is less than 10% complete drive the talk: workload-aware… | ||
| Greg Ernst, ScaleFlux | CASPA CEO Summit 2026 | Greg Ernst | 18m |
Claims that AI growth is hitting a data wall not power, with context windows and multi-user token retention far outstripping HBM and forcing use of DRAM and… | ||
| Naveen Rao, Unconventional AI | CASPA CEO Summit 2026 | Naveen Rao | 20m |
Claims a top-to-bottom computer that erases the line between hardware and model and exploits physical dynamics instead of floating‑point abstractions… | ||
| Faris Sbahi, Normal Computing | CASPA CEO Summit 2026 | Faris Sbahi | 23m |
Normal Computing claims HBM is the scaling bottleneck — it uses three to four times more wafers than DRAM and is effectively sold out through 2028 — and… | ||
| Melton Chang, Schneider Electric Power Systems | CASPA CEO Summit 2026 | Melton Chang | 18m |
GPU and semiconductor generations turn over in ~1 year while utility power cycles run 6–10 years, creating a mismatch that forces power to go distributed as AI… | ||
| Ankur Gupta, Siemens | CASPA CEO Summit 2026 | Ankur Gupta | 25m |
Claims that AI halves custom-silicon schedules — 18 months to 9 months with OpenAI and Broadcom — and that EDA must remain deterministic drive the work… | ||
| Sid Sheth, d-Matrix | CASPA CEO Summit 2026 | Sid Sheth | 13m |
Claims rack-scale is now the unit of AI inference and backs it with a multi-megawatt Corsair XPU cluster built after acquiring GigaIO and Wallaroo to move from… | ||
| David Lam, Lam Research | CASPA CEO Summit 2026 | David Lam | 19m |
Multibeam claims to compress an 80 cm electron‑beam column down to pen-sized multibeam columns and tile those pens for parallel writing so maskless… | ||
| Hong Hou, Semtech | CASPA CEO Summit 2026 | Hong Hou | 18m |
AI data centers become a team sport of hundreds of thousands of interconnected ASICs and GPUs where connectivity, latency and power rival raw compute as… | ||
| nothing matches | ||