Vinod Khosla, the venture capitalist behind Khosla Ventures and an investor in OpenAI, laid out two moats he expects will decide the personal AI market. The first is trust: a company that consumers believe will handle their most sensitive data responsibly. The second is execution: a product that actually finishes the job it is asked to do, rather than half completing it.
“I believe the moat will be trust in a company that develops trust with consumers,” Khosla said. “And that’s where Meta has a huge disadvantage.”
The clip, posted by the technology show TBPN on X on September 21, cuts off right after that line, before Khosla names his second target. Meta’s assistant efforts have leaned on its existing social graph rather than a record of guarding personal data, the gap Khosla appears to be pointing at.
For any team building a personal AI assistant, Khosla’s framing sets the real test: whether users trust the company with their data, and whether the product reliably finishes what it starts, benchmarks aside.
Per a clip posted by TBPN on X on September 21, 2026, quoting Vinod Khosla.
