The Rise of the AI-Native Rollup

Today, Reuters reported that OpenAI is in advanced talks with TPG, Advent International, Bain Capital, and Brookfield to form a $10 billion joint venture distributing its enterprise products across their portfolio companies. Anthropic is reportedly in parallel discussions with Blackstone, Hellman & Friedman, and Permira.

The two most valuable AI companies on the planet are not waiting for enterprises to come to them. They are going to private equity and asking for help getting through the door.

This tells you everything about where we are in the AI adoption cycle. And it tells you why the AI-native rollup is emerging as one of the most compelling investment theses of the decade.

The Adoption Problem

The AI industry has a dirty secret: institutional adoption is brutally slow.

Individual adoption happened fast. Getting AI embedded into the actual workflows, processes, and economics of operating businesses? That transition is barely underway.

George Sivulka, CEO of Hebbia, laid this out brilliantly: "AI just made every individual 10x more productive. No company became 10x more valuable as a result. Where did the productivity go?"

His answer draws on a striking parallel. In the 1890s, textile mills replaced steam engines with electric motors. For thirty years, output barely moved. It wasn't until the 1920s, when factories were completely redesigned around electricity, that electrification delivered returns.

We swapped the motor. We haven't redesigned the factory.

Humans Are the Bottleneck

The reason is simple: humans resist change.

Every employee now has their own ChatGPT habits, their own prompting styles, their own outputs that don't connect to anyone else's work. The technology is 10x better. The organization is running on the same operating system it had before.

Worse, the most senior people are typically the slowest to adopt. They have the most entrenched workflows and the least incentive to change.

This is why selling AI software to existing companies is so hard. You're not selling technology. You're selling organizational transformation. OpenAI knows this. That's why they partnered with McKinsey, BCG, Accenture, and Capgemini through "Frontier Alliances." They built the best AI in the world and still need consulting firms to get it adopted.

The proposed $10 billion PE joint venture is the latest escalation: if you can't sell the software into the company, go to the people who own the company and push it through from the top.

It's rational. But there's a more elegant approach.

The AI-Native Rollup: Skip the Adoption Problem Entirely

What if, instead of selling AI to an existing company and hoping someone uses it, you bought the company and deployed the AI yourself?

That's the AI-native rollup thesis. And it's already producing results.

In fragmented, people-intensive industries — insurance, accounting, legal, managed services — thousands of small businesses operate with outdated technology, aging owners, and manual processes consuming 60–80% of their cost base. These businesses are acquirable at 3–6x EBITDA. Post-acquisition, an AI-native operator deploys technology that expands margins from 15–20% to 30–40% and creates a platform that re-rates at 10–15x on exit.

The buyer doesn't need to convince anyone to adopt AI. They own the company. They deploy on Day 1. Change management isn't a sales process. It's an integration playbook.

This collapses the 30-year electrification delay into 90 days.

The Playbook Is Already in Motion

Over the past eighteen months, more than $3 billion has been deployed into AI-native rollup strategies across half a dozen verticals.

General Catalyst's "Creation Strategy" ($1.5B+ deployed) seeds AI-native holding companies across verticals. Long Lake Partners (HOA management) reportedly hit $100M in EBITDA in under two years. Crescendo (call centers) achieved 4x margin improvement by replacing 90% of human agents with AI. Eudia (legal) raised $105M with $75M earmarked for law firm acquisitions and has already acquired Johnson Hana, a 300-lawyer legal services firm.

Crete (Thrive Capital) is executing in accounting: 30+ CPA firm acquisitions, $300M+ revenue, reportedly the fastest-growing company in the space. Buy from aging owners, deploy AI across tax, audit, and advisory, build a platform no individual firm could replicate.

Gyde ($60M, Lightspeed/Optum Ventures) launched in January 2026 as an AI-native insurance brokerage. Acquire health insurance agencies, deploy GydeOS and an AI assistant called Gia to automate renewals, scheduling, onboarding, and outreach. Brokers focus on relationships. AI handles the rest.

Shield Tech (Thrive Capital, OpenAI) is rolling up managed IT services. OpenAI took an equity stake in December 2025: Shield gets frontier models, OpenAI gets operational data from real enterprise IT environments.

Why This Works Financially

AI doesn't just create operational leverage. It creates margin transformation at a speed traditional playbooks can't match.

In a traditional rollup, margins might improve 200–400 basis points over a five-year hold. In an AI-native rollup, EBITDA margins can expand 1,000–1,500 basis points within 12–18 months.

A practical example: buy a $1.5M EBITDA practice at 4x ($6M). Deploy AI. Margins expand from 20% to 32%. EBITDA grows to $2.4M without adding a dollar of revenue. At a platform exit multiple of 10x, that single acquisition is worth $24M on perhaps $2–3M of equity invested after leverage. That's 7–10x on a single tuck-in, before any revenue growth.

Now do that fifteen times in three years.

What This Means

Today's news confirms what the AI-native rollup pioneers figured out a year ago: the hardest problem in AI isn't building the model. It's getting it deployed into real businesses that generate real revenue.

OpenAI is solving this by partnering with PE firms who control the companies. The rollup operators are solving it by becoming the companies.

Both approaches acknowledge the same truth: selling software is not enough. AI only creates value when it's embedded in the operating model of a business.

The difference is that rollup operators own the upside. They don't collect a SaaS subscription. They capture the margin expansion, the multiple re-rating, and the platform premium at exit.

The factories are waiting to be redesigned. The question is who will own them when they are.

Alex Christodoulou is the CEO and Co-Founder of MaTrx.ai, an AI-native investment banking firm that powers acquisition strategies for technology-forward buyers in fragmented markets.