Menlo Ventures’ Matt Murphy explains what AI startups founders must do differently
TechCrunch Theresa Loconsolo, Julie Bort ● Covered by 3 sources
Anthropic's revenue jumped from $9B to a $47B run rate in one year. Menlo's Matt Murphy says he's never seen growth like it in 25 years of investing.
Based on reporting by TechCrunch, Theresa Loconsolo, Julie Bort — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
Matt Murphy has sat through the internet boom, the mobile shift, and the first wave of cloud computing. He says none of it compares to what he's watched happen with Anthropic. The company went from roughly $9 billion in annual revenue to a $47 billion run rate by May, and Murphy, whose firm Menlo Ventures led Anthropic's $500 million Series D, calls it the fastest ramp he's tracked in a quarter century of writing checks.
The bet didn't look obvious at the time. Menlo backed Anthropic at a $4 billion valuation when the company had no revenue and hadn't even launched a product yet, according to Murphy, and the deal was awkward enough that it didn't cleanly fit inside any of the firm's existing funds. What convinced him wasn't the model itself. It was watching Google and Amazon come in as investors early on, a signal Murphy describes as the first green shoot that this wasn't just another lab chasing benchmarks.
Murphy's real argument is that a good language model was never going to be the moat. Plenty of labs can train something impressive. What separates Anthropic, in his telling, is the layer built on top: Claude Code, the MCP protocol, and the newer Claude Skills feature, which together turned a strong model into something closer to a platform other companies build businesses around. That's a different kind of defensibility than raw capability, and it's the reason he thinks the company can keep compounding rather than getting leapfrogged by the next big benchmark score.
He also pushed back, on the podcast, against the narrative that Anthropic's recent Mythos rollout was more about marketing than actual safety work. That's a notable defense coming from an investor with a financial stake in the story, but Murphy frames it as consistent with how the company has operated from the start, not a pivot.
What sticks with Murphy now is what he's seeing further downstream. He points to startups like Lovable and Legora as growing faster than anything he's witnessed in his career, faster even than the early Anthropic curve. For founders, that's less a comforting data point and more a warning: the bar for what counts as fast growth just moved, and standing still for even a quarter can mean falling behind a company that didn't exist a year ago.
My take — AI-written commentary, not fact-checked reporting
Every VC on a winning cap table says the model wasn't the moat, right after the model made them rich, so I'd treat that framing with some skepticism. But the Lovable and Legora growth numbers are the real story buried in here — when downstream apps outgrow the foundation model companies that power them, that's the actual signal that this market has shifted from infrastructure hype to genuine product-market fit, and founders ignoring that shift are the ones who'll get left behind.
Read more about this at: TechCrunch
