Venture capital is supposedly the land of the contrarian. The bastion of the misfits and dreamers and change-makers. The home of free thinking.
…so why is everybody doing the same thing these days?
We aggregated data from over 50,000 US startups, ranging from pre-seed upstarts to pre-IPO giants. After pouring over the numbers, there was a consistent pattern that leapt off the page:
Concentration.
First: there is concentration in sectors. Well, one sector in particular. Can you guess which?
AI, AI, AI. These “AI dominance” charts are starting to bleed together but just because the message is annoyingly overplayed doesn’t make it wrong. More VC capital is flowing to AI companies across basically every single stage of venture.
There is some debate around AI-enabled vs AI-native. At least within software categories, that debate is a bit semantic. However some VCs remain more focused on “pure AI” plays like foundational models and data-tuning startups, while others are happy to place bets on vertical SaaS that simply uses the latest tech.
Second: concentration in companies.
More dollars are flowing into startups but fewer companies are actually raising rounds (at least within the early stage). So each investment is slightly more consensus.
Beyond fewer initial investments, the lead investor in each of these rounds is gobbling up more share.
Likely a couple strategies coming to a head here. On the mega-fund side, buying more share in Seed or Series A rounds makes perfect sense (what’s a few extra million anyway?). In response, smaller funds may also want higher allocations in early rounds in anticipation of getting priced out of future fundraises.
Of course, it’s not just about the first investment. Take a look at this bridge round data.
Now a bridge round is usually not a great signal. If the company needs some extra cash before the next primary round, things are not going well.
And that’s what the data shows, by and large. Companies that needed a bridge (in orange) graduate to the next round less often that companies that don’t (in black)...except in the most recent quarters, where the dynamic is flipped.
Why? Because investors are getting so damn excited about their best companies that they are straining to give them more money before the next round comes in. VCs call this “pre-empting” as opposed to “bridging” and I used to think it was BS. Today, it’s common.
I think a startup ecosystem that backs weird ideas and non-consensus founders is better for us collectively.
Third: concentration in geography.
The rumors of the demise of Silicon Valley have been greatly exaggerated. I mean, look at this ranking for the last 12 months or so:
Number 1 in every major category (the metric is total fundraising by startups in that metro area) save for a complete flop by coming in second in Biotech. Congrats Boston.
But it’s not just a total capital raised thing. The Bay Area hype cycle applies to startup valuations just as it applies to dollars raised.
The chart shows that 44% of the startups that raised in the top decile of seed valuations are in the Bay. Basically if you are only trying to optimize for the highest valuation (which is not a great idea, to be clear) you effectively have to move to SF.
So we have three separate thematic concentrations occurring at the same time. Investors are zeroing in on AI startups, investing more money in fewer companies, and doing so primarily in the Bay.
You can layer in a bonus concentration, if you like - media attention. On X, across newsletters, in old media headlines, the focus seems to be affixed permanently to the 10-15 companies that dominate the AI space. Good luck getting a funding announcement covered if you aren’t in the dominant theme.
Is this all a good thing?
Probably not. I think a startup ecosystem that backs weird ideas and non-consensus founders is better for us collectively.
But that would require us to break our current concentration.
This Week Across Topline
This Made Us Think
Elad Gil on Which AI Markets Have Winners — and Which Are Still Wide Open — I think Gil captures the weird in-between moment we’re in: part gold rush, part endgame. Some markets feel locked up; others haven’t even started forming. The takeaway isn’t who’s winning… it’s how early “winners” might not matter if the next leap resets the board again.
Seriously Consider Selling: OnlyCFO — I think this post cuts through a lot of founder wishful thinking. The exit landscape has changed. PE money isn’t what it was, and “strategic” now means “we’ll buy your partner before we build it ourselves.” The takeaway isn’t to sell fast, but to build like selling is one of your real options. Relationships, not rounds, are what de-risk you now.
Everyone Should Be Using Claude Code More — I think this post captures the inflection point for AI tools: when they stop being “tech demos” and start quietly rewiring workflows for everyone else. Claude Code feels like the first AI product built for utility. Powerful, practical, and surprisingly human in how it handles chaos.
Become a Topline insider by joining our Slack channel.
We want to hear your feedback! Let us know your thoughts.
This Made Us Think: Curated by Kyler Verney.
Imagery by Neil Topinka.