The Reverse Acquihire
Why we're obsessing about the human talent that will eliminate the need for so much human talent.
By John Brewton
One of the strangest trends in technology right now is also one of the most telling.
The largest companies in the world (Google, Meta, Amazon, Microsoft, OpenAI, Anthropic, and Perplexity) are in a fevered race to build and control the future of artificial intelligence. They spend tens of billions on infrastructure, training clusters, foundation models, inference acceleration, and distribution. They're building tools that promise to eliminate knowledge work, displace software development, and reduce the time it takes to do just about everything by orders of magnitude.
Yet, these companies are most concerned with acquiring the human assets needed to build the products they’re racing to develop.
Elite research scientists, model developers, and early-stage technical founders are being paid money historically appropriated for superstar athletes and sizable industrial conglomerates. They aren’t buying companies for their revenue, logos, branding, or traction. They are buying superstar engineers and elite teams of engineers and scientists who know how to build.
Paradoxically, the very people who are building the systems that promise to make human work obsolete are the most valuable assets in the world right now.
It’s a signal. And it might be the most hopeful one we’ve got.
Six Deals That Say Everything
In the last 18 months, we've seen a shift in how acquisitions are structured. These aren't classic M&A deals. They’re not acquihires in the old sense either. They're hybrid models, designed to be fast, flexible, and focused almost exclusively on people.
Here are a few:
Windsurf → Google (July 2025)
Google didn’t buy the company. Instead, they structured a deal where they acquired the co-founders and R&D team, and secured a non-exclusive license to the underlying tech. Windsurf remains independent. The estimated value is +$3 billion. OpenAI was initially reported to be making this purchase, until they weren’t and Google took care of business.
Scale AI → Meta (June 2025)
Meta took a minority stake, 49%, with no voting control. They secured a multiyear data services pact and got Scale’s CEO, Alexandr Wang, into the Meta fold. Valuation? $14.8 billion.
Covariant → Amazon (August 2024)
Amazon took the co-founders, 25% of staff, and a license to the core “Covariant Brain” model for ~$400 million. Covariant continues to exist independently.
Character AI → Google (August 2024)
The same playbook. Google acquired 30 engineers, including the two co-founders, and licensed Character AI’s large language model non-exclusively. The company didn’t change hands. The estimated price tag: $2.7 billion.
Adept → Amazon (June 2024)
Amazon didn’t even acquire the company. They hired most of the execs and researchers outright. Adept retains its corporate shell and a perpetual license to its tech. Payout: ~$440 million.
Inflection AI → Microsoft (March 2024)
This one was closer to a traditional deal. Microsoft formed a new internal unit (“Microsoft AI”) and paid ~$650 million to bring on co-founders Mustafa Suleyman and Karén Simonyan, plus license their foundational models.
The company doesn’t matter; the team does. The brand is irrelevant. The model is secondary. What matters most are the people who can keep building.
Code is Commoditized. Talent is Not.
This shift reveals something strategic about where value resides in the AI era.
Code can be cloned.
Models can be trained and retrained.
Data pipelines can be mimicked or licensed.
But the ability to architect a new system from scratch, optimize performance at scale, or train a breakthrough foundation model with minimal resources, those abilities still live inside people.
Google, Amazon know this. Meta and Microsoft, too.
We’re not trying to “own” companies anymore. We’re consumed by a race to get the right talent. Their focus is solely on embedding capability into their core stack. That means acquiring small, high-output teams and giving them the tools, resources, and mandates to keep building within their ecosystem.
The Irony at the Heart of the AI Boom
It’s funny, almost poetic.
The biggest players in tech are promising a future where machines can reason, write, create, and collaborate like humans. They're rolling out tools that can pass bar exams, write code in real time, generate photorealistic images, and carry on conversations indistinguishable from humans.
And yet, they can’t make it happen without humans.
It’s an essential story of the current moment because it reveals a paradox at the heart of AI: The more powerful the automation becomes, the more valuable the human architects behind it become.
The most innovative system still requires someone to define the problem.
The most powerful model still needs someone to deploy it with intent.
The most disruptive tool still needs to be operated.
A Little Hope
There’s been no shortage of fear-mongering around AI. And in fairness, some of it is justified. There are real risks ahead: economic, political, and ethical.
But these deal structures remind us it’s also about the people who make the machines work.
It’s about clarity of vision.
It’s about technical craftsmanship.
It’s about the human element.
You matter.
Your execution matters.
Your team matters.
Your ability to think clearly, build simply, and act decisively, that’s still the edge.
A Closing Reflection
Focus on the fundamentals.
Invest in your people.
Develop original capabilities.
Design teams that can build under pressure.
Protect your IP, but know your execution is your moat.
That’s what the world’s most powerful companies are really trying to buy.
Not ideas.
Not models.
Not businesses.
Just execution.
Just the right people, working in the right direction, building the future we’re all imagining.
And that means the future isn’t being written for us.
It’s still being written by us.
Thanks for reading Operating by John Brewton.
If this helped you see a clearer picture of where the world is going, consider becoming a paid subscriber. I believe my work can add real value to your work, which is always my goal.
✅ Do. Fail. Learn. Grow. Win.
✅ Repeat. Forever.
📬 Subscribe to Operating (🔗 in profile) for deep dives on the history and future of operating companies.
♻️ Restack to share with your Substack network
Thank you for taking the time to read.
- John -
John Brewton documents the history and future of operating companies at Operating by John Brewton. After selling his family’s B2B industrial distribution company in 2021, he has been helping business owners, founders and investors optimize their operations ever since. His frameworks have generated over $500M in enterprise value. He is the founder of 6A East Partners, a research and advisory firm asking the question: What is the future of companies? He still cringes at his early LinkedIn posts.