Companies Are Done Renting Their AI — Here's What That Shift Means for Your Business

The open source AI wave is no longer a fringe movement. According to Hugging Face CEO Clem Delangue, it is now mainstream enterprise strategy — and the numbers back him up.

In a recent interview with TechCrunch AI's Theresa Loconsolo, Delangue described how Hugging Face has grown into something like a GitHub for artificial intelligence, a central hub where builders share and download open models and datasets. The platform is now used by roughly half of the Fortune 500. That is not a niche developer tool. That is critical infrastructure.

And the story Delangue keeps hearing from enterprise customers? The same one, over and over: companies start by renting their AI from a big proprietary provider, then gradually decide they want to own it instead.

Why Enterprises Are Walking Away from Proprietary AI Lock-In

The "rent versus own" dynamic has played out before in enterprise software. Cloud computing, CRM platforms, and even productivity suites all went through a version of this debate. AI is now entering that same maturity phase.

For the past several years, the default move for companies adopting AI was to plug into a large language model API — OpenAI, Anthropic, Google — and build on top of it. Fast, convenient, and low friction at the start. But that approach comes with hidden costs that compound over time: usage fees that scale with growth, limited control over model behavior, data privacy concerns, and the constant vulnerability of a vendor changing pricing or deprecating a model.

Open source changes that calculus entirely. With open models available through platforms like Hugging Face, a company can download a model, fine-tune it on proprietary data, run it on its own infrastructure, and pay nothing in per-token fees. The upfront investment is real, but so is the long-term leverage.

Delangue's point is not just philosophical. It reflects a structural shift in how sophisticated organizations think about AI as a competitive asset rather than a utility subscription.

What This Means for SMBs

Here is where it gets interesting for smaller businesses and lean teams.

The Fortune 500 has the engineering headcount to spin up open model deployments and manage fine-tuning pipelines. Most SMBs do not. So does the "own don't rent" movement pass them by?

Not necessarily — but it does require a more deliberate strategy.

The practical reality is that open source AI has become dramatically more accessible. Models that would have required serious infrastructure investment two years ago can now run efficiently on modest hardware or through low-cost hosting providers. The gap between "enterprise open source AI" and "accessible open source AI" is closing fast.

For SMBs, the immediate takeaway is less about switching off proprietary APIs entirely and more about avoiding over-dependence on any single vendor. A mixed approach — using hosted APIs for general tasks while exploring open models for specific, sensitive, or high-volume workflows — offers a reasonable middle path.

The deeper strategic question is: which parts of your AI usage are core to your competitive advantage? Those are the workflows where ownership, customization, and data control matter most. Those are the areas worth exploring open alternatives. For context on evaluating your options, see our breakdown of AI tools for business and how to match them to your actual operational needs.

The Bigger Picture: AI Is Becoming Infrastructure

What Delangue is really describing is AI graduating from a novelty feature into genuine business infrastructure. And infrastructure, historically, gets owned rather than rented once it becomes essential.

This shift also has implications for how teams think about AI-powered automation. When your automation workflows depend on a model you do not control, every vendor decision they make becomes your operational risk. Open source models, deployed on your own terms, remove that dependency.

Platforms like WRRK.ai are built for this transitional moment — helping business teams put AI to work across real workflows without requiring a dedicated ML engineering team to do it.

The companies that will win this next phase of AI adoption are not necessarily the ones with the biggest budgets or the most sophisticated models. They are the ones that think clearly about what they need to control, what they can afford to outsource, and how to build systems that compound over time rather than just bill by the month.

Original reporting by Theresa Loconsolo for TechCrunch AI, published July 10, 2026. Read the full article at TechCrunch.

Frequently Asked Questions

Why are companies moving away from proprietary AI providers?

Companies are increasingly concerned about vendor lock-in, escalating API costs, data privacy, and lack of control over model behavior. Open source AI allows businesses to fine-tune models on their own data, run them on their own infrastructure, and avoid per-token fees that scale with usage. Hugging Face CEO Clem Delangue notes this transition is now a common pattern among enterprise adopters.

Is open source AI a realistic option for small businesses?

Yes, increasingly so. While large enterprises have historically had the engineering resources to deploy open models, the barrier to entry has dropped significantly. Smaller teams can now access capable open models through affordable hosting options and user-friendly platforms, making the own-versus-rent decision relevant for SMBs as well.

What is Hugging Face and why does it matter for enterprise AI strategy?

Hugging Face is a platform often described as the GitHub of AI — a central repository where developers and companies can share, download, and deploy open source AI models and datasets. With roughly half of the Fortune 500 now using it, it has become a core piece of enterprise AI infrastructure and a key driver of the open source AI movement.

Start building AI workflows your business actually owns — explore WRRK.ai today.

Companies Are Done Renting Their AI — Here's What That Shift Means for Your Business