🔮 X-raying OpenAI’s unit economics

2026-02-13 · Source: Exponential View · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Corporate Strategy & Leadership, Emerging Technologies & Innovation · Depth: Advanced, extended

Summary

A recent analysis by Exponential View and Epoch AI investigated the unit economics of frontier AI models, specifically OpenAI's GPT-5, to determine their profitability. The research, discussed in a live conversation, found that while OpenAI likely generated more revenue than the cost of compute for GPT-5, its overall operating margins were thin or negative after accounting for staff, sales, marketing, administrative costs, and the Microsoft revenue-sharing agreement. Notably, the R&D expenditure in the four months preceding GPT-5's release probably exceeded the gross profits earned during GPT-5 and GPT-5.2's entire tenure. The analysis highlights the short lifespan of preeminent models, which act as rapidly depreciating assets, and explores potential paths to profitability, including the role of advertising and the strategic focus on enterprise solutions versus consumer markets.

Key takeaway

For entrepreneurs or investors evaluating AI companies, recognize that current frontier models, despite high valuations and significant capital expenditure, operate on thin margins due to immense R&D costs and short product lifecycles. Your investment strategy should account for AI models as rapidly depreciating assets, prioritizing companies with clear paths to enterprise monetization or robust infrastructure plays over those solely reliant on consumer-facing models or speculative bets. Scrutinize gross profit margins relative to development costs to gauge long-term viability.

Key insights

Frontier AI models like GPT-5 face significant profitability challenges due to high R&D costs and short model lifespans.

Principles

• AI models are rapidly depreciating assets.
• Gross profit margins for foundation labs are lower than traditional software businesses.

Method

The methodology involved piecing together public financial information for OpenAI, projecting historical data to 2025, and breaking down costs into categories to assess realistic approximations of margins and R&D spend.

In practice

• Consider infrastructure as a more enduring asset than rapidly depreciating AI models.
• Focus on enterprise solutions for higher revenue potential in AI.

Topics

• OpenAI Unit Economics
• Frontier AI Profitability
• Large Language Models
• AI Infrastructure
• AI Market Dynamics

Best for: Entrepreneur, Investor, Director of AI/ML, Business Analyst

Related on AIssential

• Weekly Top Picks #112: Besides Moltbook — Frontier AI models are rapidly depreciating assets, with R&D costs often outpacing revenue generation before obsolescence.
• What is causing OpenAI to lose so much money compared to Google and Anthropic? — OpenAI's high costs stem from its broad strategy, infrastructure choices, and large free user base.
• Deepseek reportedly prioritizes AGI research over quick profits despite billions in funding — Deepseek prioritizes AGI and open-source development over short-term profits, even with significant state-backed funding.
• The Subsidized AGI Economy — AI labs intentionally subsidize power users to acquire critical agentic workload data and secure future market position.
• Why companies are penny-pinching on tokens — Frontier AI models offer performance gains but incur significant token costs, challenging ROI measurement for enterprises.
• AI is the Concorde of our time (Ep. 309) — Frontier AI's technical brilliance is overshadowed by its economically unsustainable, industrial-scale operational costs.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Exponential View.