At a Y Combinator event on Tuesday night, Sam Altman delivered what YC partner Tyler Bosmeny called a "mic drop moment." The OpenAI CEO offered $2 million worth of OpenAI tokens to every startup in the current YC batch in exchange for equity in those companies. Instead of cash, startups would receive an allotment of AI tokens that can be used to access OpenAI’s models for building their products.

The announcement was made via a tweet from Altman on May 20, 2026, where he expressed excitement about seeing what "tokenmaxxing startups" will achieve. Y Combinator has about 169 startups in this cohort, according to its directory.

How the deal works

The exact equity each startup gives up is not determined at signing. It will depend on the company’s valuation when it raises its first priced round—typically a Series A. YC managing director Jared Friedman confirmed to TechCrunch that the deal is structured as an "uncapped SAFE" (Simple Agreement for Future Equity). A SAFE is YC’s standard instrument for early-stage companies raising money before a formal valuation. An uncapped SAFE has no ceiling on the valuation, which benefits founders: if the conversion valuation is higher, OpenAI receives a smaller ownership percentage.

Speculation on X suggests that if a startup reaches a $100 million valuation, OpenAI might end up with roughly 2% equity. However, without seeing the actual terms, this cannot be verified. For comparison, Y Combinator itself takes 7% equity for a $500,000 cash investment in its standard deal, providing access to its powerful network.

Why OpenAI made this move

The deal serves dual purposes for OpenAI. First, it gains equity in a large crop of early-stage startups, meaning it shares in their success. Second, it incentivizes these startups to build their businesses on OpenAI’s platform rather than defaulting to competitors like Anthropic’s Claude Code. Even if the tokens don’t lock startups in permanently, they create a strong initial preference for OpenAI.

Additionally, as inference costs continue to fall, the tokens that OpenAI is giving away today may become very inexpensive to produce in the future. This makes the equity received in return look increasingly cheap over time.

Reactions from the startup community

Unsurprisingly, the offer has sparked intense debate on X. Supporters argue that the tokens help eliminate one of the biggest costs for early-stage AI startups: infrastructure bills. AI model usage can spiral quickly, consuming a disproportionate share of a startup’s limited budget. By accepting tokens instead of spending cash, startups can preserve capital.

Critics, however, are more cautious. Seed investor Jason Calacanis, who runs a competing accelerator, warned that accepting tokens from OpenAI carries a non-zero risk of the company studying a startup’s work, copying the idea, and integrating it into a free product. "This is the classic platform playbook — be careful, founders!" he posted.

The fear that Big Tech companies like OpenAI and Anthropic could absorb every promising AI startup idea is real. However, the truth is that OpenAI can already access any startup’s product and data when the startup pays for tokens. By taking an equity stake, OpenAI may have more incentive to support the startup’s success rather than undermine it. Furthermore, as former YC president and a recurring guest speaker, Altman already has considerable access to each cohort and its ideas.

The equity trade-off

The bigger issue for YC startups is whether a budget of tokens from a single AI player is worth giving up additional equity. Startups already face equity dilution from YC (7%), seed investors (often 20% or more), and employee stock grants. Adding another slice might leave founders with a smaller ownership position.

Yet the alternative is paying for tokens with cash, which is an even scarcer resource at the early stage. Bootstrapping AI infrastructure can force startups to limit experimentation or scale slower. The token deal effectively converts future cash expenses into equity, potentially aligning incentives between the startup and OpenAI.

Another danger is that a startup might burn through its $2 million token budget without achieving meaningful traction, having given up equity for nothing. But that risk exists with any investment—cash or tokens. Ultimately, the decision hinges on each startup’s confidence in its ability to create value using OpenAI’s models.

Background on Y Combinator and Safe Notes

Y Combinator is arguably the most influential startup accelerator in Silicon Valley. Since its founding in 2005, it has backed companies like Airbnb, Stripe, Dropbox, and Reddit. Its standard investment model involves accepting 7% equity in return for $500,000 and participation in a three-month program that culminates in Demo Day. The SAFE (Simple Agreement for Future Equity) was invented by YC to simplify early-stage fundraising. A SAFE is not a debt instrument; it gives investors the right to receive equity in a future priced round. An uncapped SAFE, as used here, means the valuation cap is removed, which is favorable to founders because it prevents excessive dilution if the company’s valuation skyrockets.

OpenAI, founded in 2015 as a non-profit, transitioned to a capped-profit model in 2019 and has since become the leading AI research and deployment company. Its products include GPT-4, DALL-E, and the API that many startups rely on. The token-based offer from Altman signals a new approach to corporate investment, blending venture capital with platform incentives.

Broader implications for the startup ecosystem

This deal could reshape how big tech companies invest in early-stage startups. If successful, it might encourage other AI firms like Anthropic, Google, or Microsoft to offer similar token-for-equity deals. The strategy allows these companies to nurture an ecosystem of startups that are deeply integrated with their platforms, potentially creating moats against competitors.

However, regulators might take note. If tokens become a dominant form of investment, it could raise antitrust concerns about tying usage of a platform to equity stakes. Startups may also worry about data sharing and the risk of ideas being copied. For now, the YC batch of Spring 2026 has to decide whether the tokens are worth the equity—a choice that could define the next generation of AI startups.

Source: TechCrunch News