Instinct raised $350 million at a $2.5 billion valuation, and I still can’t tell you whether the product is any good.
That’s not a knock on the team. It’s a knock on the process. I review AI tools for a living, which means I spend my days running agents against real tasks and watching them fail in interesting ways. What I’ve learned is that funding rounds and product quality are barely correlated anymore. They’ve become two separate games played by two separate audiences, and the round is the one that gets the headline.
What we actually know
Here’s the full extent of the verified picture: Instinct is a viral AI startup. It raised $350M. The valuation is $2.5B. TechCrunch reported it. The Wall Street Journal ran an exclusive describing the latest viral AI assistant moving fast through Silicon Valley.
That’s it. That’s the dataset.
Notice what’s missing. No revenue figure. No user count that’s been audited by anyone. No benchmark results. No independent evaluation of how the assistant handles the boring, unglamorous work that separates a demo from a tool — long context, ambiguous instructions, tasks that require admitting uncertainty instead of confidently making something up.
When I evaluate an agent, those are the things I test for. None of them show up in a funding announcement, because funding announcements aren’t product reviews. They’re a signal that a group of investors, working from information you and I don’t have, made a bet. That bet may be smart. It tells you approximately nothing about whether the thing works for you.
The pattern is the story
Instinct isn’t an isolated event. Look at the same news cycle:
- Higgsfield raised a $400M Series B, quadrupling its valuation in eight months to $5.4B
- Parallel Web Systems hit a $2B valuation five months after its last big raise
- Ricursive, an AI chip startup, hit a $4B valuation two months after launch
Two months. Five months. Eight months. Those aren’t product development cycles — they’re barely long enough to onboard a sales team. Valuations are re-rating faster than anyone outside the company can evaluate what changed.
I’m not calling any of these companies overvalued. I don’t have the numbers to make that claim, and neither does anyone writing confident threads about it. What I am saying is that the velocity itself has become the narrative. “Quadrupled in eight months” is doing more work in the headline than any statement about the product could.
And that creates a genuine problem for anyone trying to choose tools. The loudest signal in the market is now the one least connected to whether software solves your problem.
Why the viral part deserves scrutiny
The WSJ framing — an assistant rocketing across Silicon Valley — is worth sitting with. Silicon Valley adoption is a real signal. It’s also a narrow one. Early adopters in tech are unusually tolerant, unusually technical, and unusually willing to work around rough edges because trying the new thing is part of their job.
Tools that spread through that crowd often stall when they hit users who don’t want to prompt-engineer their way around a limitation. I’ve seen it repeatedly: strong word-of-mouth inside the bubble, then a wall of confused reviews from people who expected the tool to just work.
Whether Instinct clears that gap is an open question. It’s not one that $350M answers.
What I’d want before recommending it
If Instinct’s team wants to convert this attention into credibility with people like me, the path is straightforward and mostly unpleasant:
- Publish failure cases, not just highlight reels. Show me where the assistant breaks down and what you’re doing about it.
- Give reviewers unrestricted access. Not a curated demo environment.
- Be specific about what the tool doesn’t do. Every honest product has a list.
- Report retention, not signups. Viral means people tried it. Retention means it worked.
Until then, my verdict stands where it started. Instinct raised a large round at a large valuation from people who presumably did their homework. Congratulations to them, sincerely — raising money in this environment is still hard, and $350M is a real vote of confidence.
But a funding round is a financial event, not a product review. Treating one as the other is how you end up with a subscription to something you stopped opening after week two.
I’ll test it. Then I’ll tell you whether it’s any good. Those are different articles.
🕒 Published: