Maor Shlomo spent the first months of Base44 sleeping in two-to-three-hour blocks. He had nobody to watch the platform overnight, so he set alarms through the night to check his servers were still running. When the platform fell over under a traffic spike, those alarms meant he caught it in ten minutes instead of six hours.

That detail comes from the founder the “you don’t need a technical co-founder anymore” crowd cites most. Base44: solo founder, no venture money, product built largely with AI, sold to Wix for $80 million in cash after six months, and roughly $150 million in ARR by May 2026 under its new owner. As evidence of what one person with an AI stack can do, it holds up. As evidence that you can skip the technical co-founder, it collapses the moment you read the details.

The pitch is louder than ever

Data published this week by equity platform altshare shows nearly one in four new startups now has a single founder, almost double the share of four years ago. The AI tools driving that shift are real, and so is the velocity. The pitch writes itself: a company of one (eight staff by the end), $189K of profit in a single month, not a dollar raised.

Base44 anchors almost every version of that pitch. So it’s worth being precise about who built it and what the job involved.

The founder in the story is an engineer

Shlomo is a programmer. Before Base44 he founded Explorium, a data analytics company backed by Insight Partners that raised a $75 million Series C. The canonical case for skipping the technical founder was executed by an experienced technical founder who used AI for implementation speed while supplying the architecture, the monitoring strategy, and every operational call himself.

The alarms are the tell. Nothing in his AI stack knew the servers needed watching, what “down” would look like, or that a traffic spike at 3am was worth losing sleep over. That judgment came from years of engineering. He also built his own agents for QA and feedback triage, which is not something you do without knowing what QA is for.

Even that wasn’t enough to ship secure

A month after the acquisition, Wiz found a critical authentication bypass in Base44. Anyone holding an app’s ID, a non-secret value visible in its URL, could register a verified account on private enterprise apps and walk straight past SSO. Wix patched it within 24 hours and found no evidence of exploitation. That was the good outcome: an engineer as founder, Wix’s security resources behind him, and the platform still carried a flaw that would have let strangers into internal HR tools and PII.

Now remove the technical founder entirely. CVE-2025-48757 documents the result: researchers scanned 1,645 production apps built on Lovable and found 303 insecure endpoints across 170 of them, with row-level security gaps leaving database tables readable and writable by unauthenticated attackers. Lovable disputed the CVE on the grounds that protecting application data is each customer’s responsibility. That defence is the whole problem. The platform says security is your job, and “you” is a founder with no way to evaluate whether the AI did it.

What a diligence review will surface

Investors have adjusted faster than pitch decks have. Diligence firms now treat “how was this code produced” as a first-order question, because roughly 41% of new code is AI-generated and the gap between “it works” and “it’s well built” keeps widening. A reviewer who pulls your repo will find the missing access controls, the secrets in client code, and the shortcut that became load-bearing, whether or not you can.

None of this says stop building with AI, and none of it says a solo founder can’t win. It says stop reading Base44 as permission to raise a round with nobody who can read the code. Budget for the judgment AI doesn’t supply: someone who can threat-model the data you store, audit what the agent shipped, and tell you what a reviewer will flag before a term sheet depends on it. A technical co-founder is one way to get that. A fractional CTO or a paid pre-raise review are others. “Nobody” is not on the list.

Shlomo covered that gap with his own expertise and still sold to Wix to get the scale muscle he lacked. If your plan for the same gap is hope, your diligence review is where hope goes to die. We do that review before your investors do.

Sources