Stripe's OpenRouter price depends on what its own stock is worth

Most of the price is paid in Stripe stock ^[2], and the report does not break out the cash portion, the share count or the valuation placed on those shares. The deal's value therefore rests on a valuation of Stripe's private shares that the report does not give. The source addresses Stripe's numbers to investors watching private tech ^[13], where no daily quote checks that valuation. The source's case for paying it is that a platform sitting between developers and many AI models is a natural toll booth, and that owning one puts Stripe closer to the AI infrastructure layer ^[21]. That case may be right. Until Stripe or OpenRouter discloses the share terms, the headline figure cannot be checked.

The singularity label has three readings, and the source supplies all of them. It can be a measurement: John Collison points to AI advances and a sharp rise in new companies ^[4]. It can be packaging, which is how the source itself treats it, with Stripe's growth figures as the substance and the branding as decoration ^[12]. Or it can be a joke that stuck, since Collison acknowledged in an October 2026 interview that the label began as somewhat tongue-in-cheek ^[6]. We think the packaging reading is closest.

The date is the first puzzle. Patrick Collison floated the first quarter of 2026 as a candidate in February ^[5], and the letter fixed January 1 in mid-August ^[3][4], so the start date was picked with about seven and a half months of Stripe's own data already in hand ^[16]. The growth also predates it. Volume of $1.9 trillion in 2025 ^[8], up 34% ^[9], implies about $1.42 trillion in 2024 ^[14] and a gain near $480 billion ^[15] before the day Stripe now calls the start. First-half 2026 revenue growth runs seven points above that 34% ^[22], though revenue and volume are different measures over different periods.

The Atlas cohort is the second. The source calls it the metric worth tracking, and also a single October snapshot of early-stage revenue that can be volatile ^[19]. Cohort revenue is the number of companies times revenue per company, and the source reports an unprecedented number of solo founders ^[10] without giving cohort sizes. If the 2026 class held five times as many companies as the 2025 class, revenue per company would be flat ^[23]. According to the source, Stripe sees new companies appear and start earning before most public data catches up ^[17], so public data cannot yet confirm the cohort figure.

We are wrong if the cohort sizes come in well under five times the 2025 class and the revenue gap survives the next quarter.