Here we go again. Another quarter, another eye-watering pile of venture capital dumped into the AI furnace. Crunchbase dropped its Q1 2026 numbers in April, and honestly, they read less like a market report and more like a fever dream written by a hyperactive spreadsheet. $300 billion into startups globally in a single quarter. Of that, a staggering $242 billion — roughly 80% — went to AI companies. Let that sink in for a second. Four out of every five venture dollars on the planet are chasing the same "AI" label.
But here is where the story gets interesting, and by interesting I mean mildly concerning. That $300 billion number? It is not spread evenly. At all. According to Crunchbase, U.S.-headquartered AI startups have gobbled up nearly 88% of AI-related funding so far in 2026, or roughly $319 billion in total. The rest of the world is fighting over the scraps, and the math is not exactly subtle about it.
The latest cherry on top came on July 1, when Together AI — the neocloud darling that rents out NVIDIA GPU clusters and hosts open-source models — announced an $800 million Series C at an $8.3 billion valuation. Led by Aramco Ventures, the round more than doubled Together AI's previous valuation. On the same day, TwelveLabs walked away with $100 million in Series B funding for its video intelligence platform. Two deals, one day, nearly a billion dollars combined, and both companies are headquartered within a 10-mile radius in San Francisco.
The Numbers Behind the Hype
Let us zoom out and look at the landscape without the rose-tinted glasses. The Digital Applied research team crunched the Q1 data and found something worth chewing on: the top four AI deals alone accounted for 65% of total AI venture funding. That is not a diversified market. That is a funnel with four companies at the bottom catching most of the rain while everyone else stands in the dry.
To put it in perspective: non-AI startups globally are competing for a pool of around $58 billion per quarter. Meanwhile, a handful of American AI companies are walking away with more capital in a single funding round than some entire sectors see in a year. It raises a question that nobody in the Silicon Valley echo chamber seems eager to answer — are we funding genuine innovation, or are we watching the biggest herd mentality trade of the decade?
Follow the Money, Question the Story
Together AI's pitch is solid on paper. They are positioning themselves as the neutral infrastructure layer for open-source AI — rent GPU time, run models, avoid vendor lock-in with the big cloud providers. It is a compelling narrative, and Aramco Ventures clearly bought it. But $8.3 billion for a company that is fundamentally a GPU rental service with some software sprinkled on top? That valuation assumes a future where enterprise AI workloads explode to a degree that makes today's cloud boom look like a neighborhood lemonade stand.
Then there is the geographical concentration. When nearly 90% of AI venture dollars are flowing into one country, you have to wonder about the structural consequences. Europe, Asia, Africa — these are not backwaters without technical talent. They are producing world-class research, founding capable startups, and building real products. But the money is simply not following. The Crunchbase data suggests that investors are not evaluating deals on merit so much as on geography. If your company is not in the Bay Area, your fundraising ceiling drops exponentially.
The Warning Signs Nobody Wants to Discuss
Let me point out three things that should give anyone pause:
- Concentration risk: When 65% of funding goes to four companies, a stumble in any one of them ripples through the entire ecosystem. We have seen this movie before — it was called the dot-com bust, and the set design looks uncomfortably familiar.
- Revenue vs. valuation decoupling: Many well-funded AI startups are burning cash at rates that would make a 2021 fintech CEO blush. Valuations are being set by competitive FOMO among investors, not by fundamentals. When every major VC firm feels compelled to have an "AI play" in their portfolio, pricing discipline goes out the window.
- The talent trap: With so much capital concentrated in the US, top AI researchers globally face an impossible choice — relocate to America or watch their careers plateau. This creates a self-reinforcing cycle that drains talent from everywhere else and inflates compensation to unsustainable levels in the Bay Area.
So, Bubble or Bet?
The honest answer is: it is probably both. There is genuine technological progress happening under all this capital. Models are getting better. Products are shipping. Enterprise adoption is accelerating. But the current funding environment feels less like a measured investment thesis and more like a collective psychosis where nobody wants to be the first to blink.
Together AI's $800 million raise will probably look prescient if AI compute demand keeps growing at its current trajectory. Or it will look like a monument to peak hype if the market corrects and GPU rental margins compress. The same logic applies to the broader $242 billion question mark hanging over the entire sector.
What is undeniable is that the concentration of AI funding into US companies — and specifically into a tiny handful of mega-deals — is reshaping the global technology landscape in ways we are only beginning to understand. Whether that concentration produces a generation of dominant companies or a spectacular implosion depends on whether these startups can turn capital into durable revenue before the music stops.
Right now, the music is still playing. But the smart money — yes, even at $319 billion — might want to keep an eye on the exit signs.
Comments