Chamath is making one of the most important business arguments of 2026.
Milk Road AI Twitter · Milk Road AI (@MilkRoadAI) · 2026-07-04
Chamath Palihapitiya argues that enterprises adopting frontier AI models are unknowingly surrendering proprietary data to third-party servers while paying up to 65x more than open-source alternatives that deliver comparable performance on most tasks.
Appears in
Extraction
Topics: enterprise-aidata-sovereigntyopen-source-modelsai-cost-analysisai-business-strategy
Claims
- Half of large US companies currently cannot generate returns that exceed their cost of capital, which has normalized to 8-11%.
- Every enterprise AI query transmits proprietary workflows, customer data, and pricing logic through external third-party servers.
- Deutsche Bank calculated a roughly 65x cost gap between frontier models like Claude Fable 5 (~$3.25/task) and open-source alternatives (~$0.05/task).
- Microsoft blocked internal employee use of Claude Fable 5 over its 30-day data retention policy.
- Open-weight models now match closed frontier systems on core agent tasks at roughly one-tenth the cost, with a high-volume deployment costing $250/day on Claude running at $12/day on an open-source equivalent.
Key quotes
companies are spending on tokens while handing over the exact proprietary advantages that make their business worth owning
Deutsche Bank calculated a roughly 65x cost gap between frontier models like Claude Fable 5 at ~$3.25 per task and open-source alternatives at ~$0.05.
a high-volume deployment that costs $250/day on Claude runs at $12/day on an open-source equivalent.