The Unsustainable Subsidy

· Source: Tomasz Tunguz · Field: Finance & Economics — Economic Analysis & Policy, Capital Markets & Investment Management · Depth: Fundamental Awareness, quick

Summary

An analysis of AI model pricing reveals distinct strategies among major vendors. Google's Gemini 3.1 Pro, despite tripling in price annually, remains the low-cost option. Its current rates are \$2.00 per 1M input and \$12.00 per 1M output. OpenAI's flagship GPT-5.5 saw prices rise after an initial subsidy period. It now costs \$5.00 per 1M input and \$30.00 per 1M output. Anthropic's Claude Opus 4.7 has maintained stable pricing, with decreases for its most powerful models. It is priced at \$5.00 per 1M input and \$25.00 per 1M output. These varied adjustments suggest a strategic shift. Vendors are moving from prioritizing market share when cash is plentiful to focusing on margins as capital expenditure sets new records.

Key takeaway

For AI Product Managers evaluating model providers, current pricing trends signal a shift towards margin-focused strategies. You should factor in Google's annual price increases and OpenAI's rising costs when projecting long-term budgets and ROI. Anthropic's stable, decreasing prices for powerful models offer a potential alternative. Reassess your total cost of ownership. High capital expenditure will likely sustain these margin-driven pricing models across the industry.

Key insights

The AI model market is shifting from growth-focused subsidies to margin-driven pricing due to high capex.

Principles

In practice

Topics

Best for: CTO, VP of Engineering/Data, Product Manager, Director of AI/ML, AI Product Manager, Investor

Related on AIssential

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Tomasz Tunguz.