Benchmark's partners acknowledge AI is straining venture capital's rule against backing rivals

A founder with an observer on the board has a short list to check on Monday: who holds an observer seat or an adviser title, and which investors also back a company next door.

Founders tend to tell themselves the old rule holds, that a firm does not back two companies that compete head-on ^[1]. What Benchmark's Eric Vishria said at Disrupt is closer to practice. Companies pivot and change, he said, and "you are going to get conflicts, and that's just the reality of the current dynamic." ^[5]

We think the evidence supports strain, not collapse. TechCrunch wrote that Benchmark is not declaring the pact dead ^[13]. Vishria said exclusivity remains "really important" at the very early stage the firm favors. "But things are evolving. There's no doubt about that," he said. ^[7]

Chetan Puttagunta said AI is accelerating the problem. As the cost of building software "effectively goes to zero," founders are "incredibly ambitious," and because they now see a "surface area" they want to claim all of it. "[i]t's probably going to happen more, where you start to run into each other," he said. ^[8]

Benchmark's reputation rests on 10-to-15-year partnerships with founders at the earliest stage ^[14]. Its Peter Fenton called the conflict issue "kryptonite" and the firm's toughest problem ^[10]. A founder should not assume a looser investor has solved what that firm has not.

The Factory dispute is narrower than it looks. Degnan announced he was joining Cognition as chief revenue officer soon after ^[18]. The TechCrunch account does not say what he was shown at Factory or whether anything left the room. A founder cannot answer either question from outside, so we'd settle what an observer sees before an argument starts, when nobody has to prove anything.

Our recommendation is to set observer and adviser access person by person, in writing, and to accept the cost: a heavier board process and investors who may feel trusted less.

Sort each person who sees board material on two axes. The first is how likely their other interests are to reach your market. The second is whether they receive roadmap and pricing detail or only financial summaries. Low overlap and summaries only: leave it alone. Low overlap with detail: get confidentiality terms in writing. High overlap with summaries only: ask the investor how it would handle a portfolio company moving into your market. High overlap with detail: cut the detail or the seat. For any name on the list, write down what that person would know on their first day if they joined your closest rival next month.