Apple's appenz: agent swarms are filling in NVIDIA's CUDA moat
appenz · x · 2026-11-23
Apple executive @appenz argues that agent swarms have essentially filled NVIDIA's CUDA moat, with two reasons: fewer model types plus agents make optimized kernels easier to build, and there's a growing trend of agents writing machine code directly, skipping CUDA altogether. He notes NVIDIA has other moats, but none as deep as this one was.
More from AGI Musings
- OpenAI Discloses 5 Years of Perplexity Data in Bid for Shared AI Safety Baseline — markchen90 · 2026-11-23
- Writer Predicts LLMs Will Train on Bot Output as Human Writing Dwindles — TuhinChakr · 2026-11-23
- What comes after know-how? LeCun weighs in on AI redefining intelligence — alexbilz · 2026-11-23
- METR-fit extrapolation implies 66-hour task horizons for 6.2 Astra, 128 hours for internal model — eliebakouch · 2026-11-23
- Uber burned its annual AI budget in 4 months with no visible improvement — yaroslavvb · 2026-11-23
- Post-training may exceed 75% of FLOPs, undercutting perplexity-based safety bars — stochasticchasm · 2026-11-23