← Home
Stratechery · May 15, 2026 · 00:25:57

(Preview) Inference in the Agentic Future, xAI Is Two Companies in One, Q&A on Elon's Lawsuit, Intel, Apple

Andrew Doyle, Ben Thompson
Show Notes

Ben Thompson and Andrew Doyle analyze the shift from training to inference compute as a market inflection point, examining Cerebras's differentiation strategy, why latency may matter less than throughput for agentic workloads, and implications for Nvidia and China's AI ecosystem. They explore xAI's dual-company structure — Colossus as training asset versus commercial inference product — and what the Anthropic-xAI compute partnership signals about market consolidation. The episode covers Elon Musk's OpenAI lawsuit strategy, Apple's gross margin expansion, and the founder-mode moment emerging from Claude access.

This summary was generated from show notes and public descriptions, not from a full transcript review. Details may contain inaccuracies.

Curious

•
Inference Compute as Market Inflection Point
The AI market is shifting from training compute scarcity to inference compute bottleneck, fundamentally reshaping the chip market and what 'speed' means for neural networks.
ClaudeChatGPTGemini
•
Cerebras Wins by Solving a Different Problem
Cerebras differentiates not by building faster chips, but by solving inference efficiency and serving customers with different compute needs than Nvidia targets.
ClaudeChatGPTGemini

Novel

•
Apple's Gross Margin Expansion as AI Land Grab
Apple's rising gross margins may not be driven by services migration, but by cornering the market for on-device inference compute and locking in a proprietary advantage.
ClaudeChatGPTGemini

Highlights

•
xAI as Two Companies in One
xAI operates as two distinct economic entities: Colossus as a training compute asset (capital-intensive, one-time), and the commercial inference product (operational, ongoing revenue).
ClaudeChatGPTGemini

Editorial

•
Market Signal: xAI Is Not the Scaling Winner
The Anthropic-xAI partnership and broader market movements signal investor skepticism about xAI's path to dominance in the AI race.
ClaudeChatGPTGemini

Misc

✧xAI being two distinct companies (Colossus training vs. inference product) is a useful framing for capital allocation
✧Inference compute becoming the bottleneck shifts the entire chip market from GPU scarcity to throughput efficiency
✧Market is signaling that Elon Musk and xAI are not the scaling winners in AI race
✧Apple's gross margin story may be less about services migration and more about the AI compute advantage being a land grab
Was this useful?