Back Rohan Paul

Rohan Paul· @rohanpaul_ai · X· 2026-10-30 09:28 · Original publication time AI score53

Anthropic S-1 shows at least two-thirds of chip spending goes to training and researchMachine translation

Automatically verified and published · Generated and evidence-checked automatically; not reviewed by a human.

AI introduction

Anthropic’s S-1 shows that at least two-thirds of its chip spending goes to model training, research and internal use of Claude. In the first half of 2026, the company spent $8.95 billion on compute, with $2.38 billion in costs to serve paying customers. The material also says Anthropic buys most capacity 12 to 18 months ahead, while revenue per compute dollar rose from $0.40 in Q1 2025 to $2.08 in Q2 2026.

Original post · Original

Rohan Paul@rohanpaul_ai

Anthropic’s S-1 shows that research and development (R&D), not serving customers, uses most of the company’s compute: at least two-thirds of its chip spending goes to model training, research and internal use of Claude.

$8.95B on compute in the first half of 2026, but only $2.38B of its costs went to serving paying customers

Serving paying customers takes barely a quarter of that compute and already brings in far more revenue than it costs.

Those customers cost $2.38B to serve against $16.29B in revenue, leaving 85 cents per dollar, up from 12 cents a year earlier.

Because Anthropic buys most capacity 12 to 18 months ahead, surging demand lifted revenue per compute dollar from $0.40 in Q1 2025 to $2.08 in Q2 2026.

09:28 · 2026-10-30 · UTC+8

Found an error? Send a correction