The cofounder prenup is a governance document, not a trust exercise

Business Insider asked three sets of AI startup cofounders what they would put in an ideal "cofounder prenup," and the answers came back as a documentation problem rather than a chemistry one ^[1]. The list from one team is equity split, vesting schedule, IP ownership, roles and responsibilities, how major decisions get made together, and what happens if someone leaves ^[2] - which is to say, the things that get called a personality clash two years in are usually unwritten terms surfacing at the worst possible moment.

The volume argument for caring about this is real. New business applications in the US were up by more than 450,000 in the first half of 2026 compared with the same period a year earlier, according to the Census Bureau figures cited by Business Insider ^[3]. Business Insider also argues that AI has raised the stakes on these early decisions, because teams are smaller, companies move faster, and pressure is more concentrated ^[4]. On a two-person company, one founder's ambiguity is fifty percent of the cap table's ambiguity.

David Emelianov, a 32-year-old engineer in the Bay Area who sold the app Trimbox with cofounder Jordan Gaston for $4 million at the end of 2025, told Business Insider there are three conversations to have before writing a single line of code: what company you are actually building, what commitment everyone expects, and what happens if someone wants out ^[5][6]. His first point is a strategy fork, not a values statement: venture-backed and swinging for the fences, or bootstrapped and cash-generating, which he calls very different games with no right answer ^[7]. He describes the failure mode plainly - one founder optimizing for a $20 million acquisition while the other wants to build a 30-year company ^[8]. Worth noting that his own realised outcome, $4 million, was a fifth of the number he uses as the mismatch example ^[19].

His second point is the one most agreements skip. Emelianov says he works on multiple projects at once and thinks it makes him a better entrepreneur, where earlier in his career he would have assumed total focus on one company ^[9]. What matters, he says, is not that everyone works the same way but that everyone agrees what "committed" means, including side projects, consulting, angel investing, and starting another company ^[10]. If forced to pick one conversation, he picks the first: confirm you are playing the same game before it begins ^[11].

Shalini Aggarwal, a 50-year-old CEO in San Jose, and Andy Ratsirarson, a 37-year-old CTO, left Amazon at different times and reconnected as cofounders of the AI startup Tenfali ^[12]. Their stress test is specific: an acquisition offer arrives, one founder wants to keep going, the other does not ^[13]. The point, they say, is not to assume the relationship fails but to make the hard conversations explicit before emotions and money are involved ^[14]. They also want a shared framework for resolving disputes, because conflict is inevitable when customers convert slowly, the product needs to change, or money gets tight ^[15]. Their enforcement clause is cultural rather than legal: once both founders align on a decision, there is no blame game later, and both own the outcome ^[16]. These talks are uncomfortable early and harder later ^[17].

The third team's section is headed "Decide whose call is final," with Praneet Dutta, a 32-year-old CEO and cofounder, though the material available cuts off before his answer ^[18]. That heading is the unresolved question in most founder agreements: not how to split, but who breaks a tie.

What to watch: whether the "committed" definition survives its first real test, which is one founder's side project growing faster than the company ^[10]; whether decision-rights language holds when an acquisition offer splits the founders ^[13]; and whether a tiebreaker is actually named rather than implied ^[18].