Inside OpenAI’s $2M “Compute for Equity” Deal With Y Combinator Startups

A new kind of startup investment: compute instead of cash

In May 2026, OpenAI CEO Sam Altman offered every startup in the current Y Combinator (YC) batch $2 million worth of OpenAI API tokens in exchange for equity. The investment is not cash; it’s usage credits that startups can spend on OpenAI models to build and run their products.

The structure uses a SAFE (Simple Agreement for Future Equity) that converts into shares when the startup raises its next priced funding round. Because the SAFE is uncapped, the exact ownership OpenAI receives will be determined later based on that future valuation.

This structure effectively turns AI infrastructure into venture capital—sometimes described as “compute for equity.”

How the token‑for‑equity deal works

The mechanics are relatively straightforward:

• OpenAI provides $2 million in API credits usable on its models and infrastructure.
• In exchange, the startup signs a SAFE agreement promising future equity.
• The SAFE converts when the company raises its next priced funding round.
• Because the SAFE is uncapped, it converts at whatever valuation that round sets rather than at a pre‑agreed cap.

Unlike typical startup credits programs, this arrangement is structured as an investment, meaning founders give up real equity rather than simply receiving promotional credits.

YC’s current batch includes roughly 169 startups, meaning OpenAI could potentially invest in the entire cohort simultaneously.

What equity stake might OpenAI receive?

The final ownership depends entirely on the valuation of the startup’s next financing round. A simplified approximation is:

OpenAI ownership ≈ $2M ÷ next round post‑money valuation.

Illustrative outcomes might look like this:

• $20M valuation → roughly 10% ownership
• $40M valuation → roughly 5%
• $100M valuation → roughly 2%
• $200M valuation → roughly 1%

Actual stakes may vary depending on additional SAFE terms, dilution from option pools, or discounts included in the agreement.

How this interacts with the standard YC deal

YC already invests $500,000 in each startup on standard terms. The structure typically includes:

• $125,000 for 7% equity, and
• $375,000 via an uncapped SAFE with a Most Favored Nation clause that converts in a future round.

That means OpenAI’s SAFE would sit on top of YC’s existing ownership structure, potentially increasing total early dilution for founders.

Why OpenAI would invest tokens instead of cash

The strategy serves several goals for OpenAI.

1. Building platform dominance

Providing millions of dollars in credits strongly encourages startups to build their products on OpenAI’s models. If those startups succeed, they may become long‑term, high‑volume API customers.

2. Lower-cost venture exposure

The marginal cost of providing compute credits can be lower than writing equivalent cash checks. This lets OpenAI invest across an entire cohort of companies while still receiving potential equity upside.

3. Expanding the developer ecosystem

By embedding its models in many early‑stage products, OpenAI strengthens its ecosystem and competes more directly with rival AI platforms from companies like Google or Anthropic.

4. Capturing upside from the AI startup wave

Instead of trying to pick a few winners, the company gains exposure to dozens or hundreds of AI‑native startups simultaneously.

Potential benefits for startups

For founders building AI-heavy products, the offer can be significant.

Lower infrastructure costs
AI inference and experimentation can be expensive. $2 million in credits can fund substantial product development and early user growth.

Longer runway
If AI usage is a major expense, credits can preserve scarce cash for hiring, marketing, or operations.

Faster experimentation
Easy access to model capacity can accelerate iteration on prompts, agents, evaluations, and workflows.

Strategic signaling
Having OpenAI as a SAFE investor could signal alignment with a major AI platform.

Key risks founders should consider

The deal also introduces trade‑offs.

Dilution risk
Because the SAFE converts later, founders may not know the exact equity cost until their next funding round.

Credits are not cash
API tokens can only be used for OpenAI services—they cannot fund salaries, legal fees, marketing, or other operating costs.

Vendor lock‑in
Building deeply around one provider’s APIs can make switching to competitors or open‑source models difficult.

Strategic overlap concerns
Some founders worry that platform providers may gain visibility into emerging product categories or usage patterns, potentially creating competitive tensions later.

The bigger picture

OpenAI’s offer represents a shift in how AI platforms compete for startups. Instead of simply selling infrastructure, the company is directly investing its own compute resources to shape the next generation of AI products.

If the model works, “compute for equity” could become a common playbook across the AI industry—where infrastructure providers act not just as suppliers, but as early investors in the companies building on top of them.