Who really controls our future with AI?

Elon Musk has a theory about the next 10 years: that it is "inevitable" that AI will surpass human intelligence and humans. There is another point of view. Israeli historian and best-selling author Yuval Noah Harari agrees with him that AI will surpass human intelligence — but he won't grant that this is all "inevitable."

People who are leading the AI revolution, Harari argues, have every incentive to describe it as fate rather than choice because it relieves them from any responsibility. When people call the AI future inevitable, they often skip over the systems making it feel inevitable: concentrated talent, capital, and decision-making. The decisions are heavily concentrated in a handful of these foundational labs, most based in Silicon Valley.

The thing is: centralized power — of any kind — poses a real risk. And while AI should continue to evolve and serve humanity, it's time we start talking more about who gets a say (and a hand) in this future.

But there's a way to change that course. It starts by looking at markets outside of the major startup hubs in the US and China, which we call Elsewhere.

1. Concentration follows the belief that concentration works

When Y Combinator moved its headquarters 38 miles from Mountain View to San Francisco, CEO Garry Tan explained why by saying “you sort of have to be in San Francisco.” He pointed to data showing that SF-based startups outperform their peers, and on paper, he's not wrong.

But every founder who hears "you have to be there" and relocates adds one more data point to the case for the next founder to do the same. Success clusters, so everyone moves toward the cluster, so success clusters more. You've got yourself a self-fulfilling prophecy.

The capital tells the same story from another angle. Foundational AI hit $178B in funding across only 24 deals by the end of Q1 2026, compared to $88.9B across 66 deals in all of 2025. In the same quarter, about 65% of all global venture investment went to just four American companies: OpenAI, Anthropic, xAI (before it became part of SpaceX), and Waymo.

In this race, the pattern is that fewer companies, fewer founders, and fewer people will be in the room, especially as “winners” pull ahead. But concentration is a belief that gets reinforced every time someone acts on it, which means it can be unwound the same way it is built: by enough people choosing not to.

2. Mega-rounds are overshadowing the meaty stuff

The track record for Elsewhere is solid. Over the past two decades, investment grew 40 times over, with VC dollars in Elsewhere rising from 10% of the global market to 60% by 2024. It was only last year, in 2025, that the trend reversed, with 64% of VC dollars going to US companies. In the first quarter of 2026, we hit “peak San Francisco:” Eighty-three percent of VC dollars went to US companies.

Those numbers, however, encompass the full spectrum of global VC. And in late-stage, private markets have been dealing with amounts we previously only encountered in the public market — i.e., until this year, the largest tech IPO in history in terms of capital raised was Alibaba back in 2014, at $21.8B. In 2026 alone, xAI raised $20B in January, Anthropic raised $30B in February, and OpenAI raised a whopping $122B in March.

But if you take out the mega-rounds, funding Elsewhere still accounts for 45% of the global venture market. Roughly half the world's venture capital still chooses to back innovation outside the traditional centers. Recent examples from our co-investment fund Endeavor Catalyst include Nigerian fintech Moove, which raised a $250M Series C round; Spain's PLD Space and its $125M round extension; Ukrainian edtech Preply's $150M Series D; Venezuelan Cashea raising $100M across two rounds; and Colorado's Maybell Quantum's $40M Series B; among many others.

3. Diversification is good for everyone

There's nothing wrong with buying Anthropic, but most investors would agree it is unwise to put all your eggs in one basket.

As makers of the tech ecosystem, we should build and bet on multiple futures. Elsewhere isn't just a bet on a founder's mission to improve their country's or region's reality; it can also be a strategic advantage. Different places produce different problems, perspectives, and ways of solving them. Those differences create opportunities that concentration can make easy to miss. For instance:

• AI voice platform ElevenLabs, although originally from Poland, is working with the Greek government to protect local dialects. It is also valued at $11B and brought our fund Endeavor Catalyst a 3X return in less than two years.
• Replit, the $9B “vibe-coding” platform founded by Palestinian Endeavor Entrepreneurs in Silicon Valley, organizes hackathons all over the world, leveraging its own product to encourage young entrepreneurs to solve real business or community challenges.
• Cashea was only able to provide access to credit to half of Venezuela's adult population — and raise $100M in back-to-back rounds this year — because their founders had lived through the very problems they set out to solve.

The more places we look, the more problems we see. The more problems we see, the more ways we can imagine solving them. And an investor with exposure to more geographies, people, and ideas has more ways to participate when one of those ideas becomes consequential. When the stakes are as high as they are with AI, having more possible futures may be one of the most valuable forms of resilience we can build.

If Musk and Harari can't agree on whether this future is something we're choosing or simply watching happen, the rest of us should stop assuming it's settled. The pull toward concentration is powerful, but a breakwater doesn't resist the ocean with a single stone. It holds because thousands of pieces, joined together, become a force the wave cannot simply wash away.

This month, our global investment fund Endeavor Catalyst took an important step in that direction, announcing its fifth and largest fund yet, with $320M in fresh capital. As a co-investment fund, Catalyst invests in 10% of rounds from Endeavor companies. That means this $320M will attract at least $3.2B to more founders from more places, with different world visions.

In 2026, we still have decisions to make about where we build, where we invest, and whose technology we put into the world. We may not control the AI wave, we may not even want to. But we can still keep it from completely reshaping the shore.

Our editorial leadLaís Grilletti wanted an authentic Italian experience. What she found was a chef with no menu and only a few ingredients available, but with a whole lot to offer.

In Italy, people pronounce my last name better than I do. I’m a Brazilian who grew up eating pasta every Sunday, and traveling to Turin felt like visiting distant family.

I’d spent the day with our filmmaker Kieran, interviewing Endeavor Entrepreneurs. By the time we finished work at 8 pm, we were very ready for dinner. A dear friend from Endeavor Italy’s team gave us the rule: If an osteria charges more than €20 for a meal, it’s probably not the real thing. We found a little place that fit the bill, where we seemed to be the only English speakers.

We asked the owner if this was a family business. “No, it’s more of a friends’ business,” he said, pointing to a photo of all the staff together on the wall. Something built with love could only serve more love.

We had to wait a little longer to order, because the owner — also our chef and waiter — had been out back writing the menu in English for us on a battered sheet of paper. Several dishes had been crossed out. Others, we discovered as we ordered, should have been.

Of course they accept cards! Just not Amex. Yes, that dish is great! No, they don’t have it. What do they have? He checked the kitchen. He had onions, cheese, and tomatoes. He could make us pasta.

Then, the dish arrived. A simple tagliolini in a slow-cooked tomato sauce, with a little parmesan on top, the kind you just roll around the fork and pray does not spray everywhere and stain your shirt. It was the best pasta of our lives.

Somewhere between laughing with the owner and walking out with happy, full stomachs, I realized that he was onto something. Friends had built a business together, the food was delicious, and we felt welcome. They knew what mattered.

From inside our tabs, headphones, and group chats, these are the signals shaping how we see the world right now.

➡️ Ask who gets to build what’s next. With Endeavor Catalyst’s new $320M Fund V, Managing Partner Allen Taylor reflects on why it matters that the next technologies shaping our lives are built from more places, by more kinds of founders.

➡️ Prioritize discovery, not disruption. Demis Hassabis from DeepMind shares how a childhood in chess and video game design led him toward AI and scientific exploration, reminding us that the most interesting journeys rarely follow a straight line.

➡️ Turn pixels into a model of reality. Packy McCormick and Runway’s co-CEO and Endeavor Entrepreneur Anastasis Germanidis make the case for AI systems that learn how the world works through video, with implications far beyond generating clips.

➡️ Steal a page from Nubank’s next chapter. David Vélez’s fintech has 140 million customers — and that’s still not enough. In this podcast, he shares how they're taking on the US and why he believes the $10T banking industry is still in the earliest stages of disruption.

Copyright © 2026 Endeavor Global, Inc. All rights reserved.

Want to change how you receive emails from Endeavor?

Endeavor Global, Inc., 28 Liberty Street, 53rd Floor, New York,New York,10005,