One Bold Bet on Anthropic Just Made a 50-Year-Old Investor Its Biggest Fund Ever
In 2024, Menlo Ventures made a 'bet-the-firm' move on Anthropic when it was still an underdog. That $500 million stake is now worth around $14 billion — and it just helped raise the firm's largest fund in half a century. A story about conviction and timing.
Here’s a story about what one well-timed bet can do. The investment firm Menlo Ventures just raised $3 billion — the biggest fund in its 50-year history — and the reason traces back to a single decision made when the outcome was anything but certain.
A bit of background for anyone outside the startup world. Venture firms raise money from big investors (pension funds, universities), then place bets on young companies hoping a few become huge. In 2024, Menlo did something unusual: it raised a dedicated $500 million pot specifically to invest in Anthropic — at a time when Anthropic had little revenue and was widely seen as the well-funded underdog to OpenAI’s ChatGPT. One partner later called it a “bet-the-firm moment,” the kind of wager that either makes your reputation or ends it.
It made their reputation. By mid-2026, with Anthropic valued above $900 billion, Menlo’s stake is worth close to $14 billion — one of the great venture returns in the industry’s history. That track record is exactly what the big investors want to be near right now, which is how Menlo raised $3 billion so easily. There’s a lesson tucked inside, though, and it’s not “make giant bets.” It’s about when the money showed up: Menlo backed Anthropic before the profits, before the proof, when belief was the only evidence available. That kind of early, patient, conviction-driven capital is far rarer — and far more valuable to a founder — than the flood of money that arrives once a company is obviously winning.
One honest caveat: paper valuations aren’t cash. Anthropic at $900 billion is a number based on optimistic assumptions about future growth, and AI valuations across the board are running hot. If those assumptions hold, Menlo looks like a genius; if they cool, the story reads differently. That uncertainty is the nature of the game, not a footnote to it.
What this means for you: If you’re building something with AI, the signal here is that money for AI-native products and tools is plentiful and hunting for a home. But the more durable takeaway is about the kind of backing worth seeking: not the biggest check, but the one that shows up early and stays aligned with what you’re actually trying to build. For everyone else, it’s a vivid illustration of how fast — and how speculatively — AI valuations have moved in just two years.
Sources
Even Zuckerberg Admits It: AI 'Agents' Are Harder to Build Than Promised
In a leaked internal meeting, Mark Zuckerberg admitted Meta's push to build AI 'agents' has stalled — despite a $145 billion bet and a painful reorganization. A useful reality check on one of the year's biggest buzzwords.
