This Fourth of July, somewhere between the fireworks and the second helping of potato salad, most of us will say the words “land of the free” without giving them a second thought. They are part of the furniture of the holiday, said so often they stop registering as an idea and start functioning as background noise. As America closes in on 250 years, it seems worth pausing on that word a little longer. Not as a political statement, and not as an argument for one flag or another in the culture wars. As a design principle. Because whether we have thought about it this way or not, freedom has been one of the load-bearing walls of the modern technology industry, and it is worth understanding why before we let it quietly get remodeled out of the building.

I first heard the phrase that gives this piece its title from my friend Ron Gula, years ago, in one of those conversations that start about a specific product and end up somewhere much bigger. We were talking about open source, and Ron said something that has stuck with me ever since: Free as in freedom, not free as in beer. It sounds like wordplay. It is actually one of the more useful distinctions in all of computing.

The Revolution was Never the Price Tag

Free as in beer means the thing costs nothing. You show up, somebody hands you a drink, and that is the entire transaction. Free as in freedom means something else entirely. It means you can see how the thing is made. You can take it apart, understand it, change it to fit your problem instead of the vendor’s roadmap, and pass your improvements along to the next person who needs them. One kind of free saves you a few dollars. The other kind of free changes what you are capable of building.

That distinction is the quiet engine behind a huge share of the enterprise technology we now take for granted. Linux did not take over the data center because it was cheap, though it was. It took over because engineers could open the hood. Kubernetes did not become the default way the industry orchestrates containers because Google gave it away out of generosity. It won because a community could shape it, extend it, and trust that no single company held a kill switch. PostgreSQL, Python, the entire stack most digital businesses run on today, all of it grew out of the same premise: that permissionless innovation beats permission-required innovation, over almost any time horizon that matters.

Enterprises that adopted open source in the 2000s and 2010s were not simply chasing a lower line item in the IT budget, though CFOs certainly noticed. They were buying insurance against being permanently at the mercy of one vendor’s pricing team. They could inspect the code. They could fork it if a relationship went sideways. They could contribute a fix upstream instead of waiting eighteen months for a support ticket to clear. Free as in beer was a nice bonus. Free as in freedom was the actual revolution, and it is the reason open ecosystems still anchor most modern architectures, even the ones built primarily on commercial software.

Then The Economics Changed Shape

Artificial intelligence is rewriting that arrangement, not through some grand ideological shift, but through a much more mundane mechanism: the meter is always running. Software used to be something you bought once, installed, and depreciated over a few years. Increasingly, it is something you consume continuously, one inference at a time. Every prompt costs something. Every agent workflow, every autonomous loop, every experiment a curious developer decides to try after hours, carries a visible, measurable price in tokens.

That sounds like a footnote about billing. It is not. When experimentation is free, or close enough to free that nobody thinks about it, organizations tend to encourage exploration. Try the weird idea. Spin up the prototype. See what happens. When every one of those same explorations shows up on a dashboard with a dollar figure attached, the instinct shifts, often without anyone deciding it should. Finance starts asking questions. Managers start pre-approving experiments instead of letting them happen organically. Innovation does not disappear in this world. It just gets more cautious, one budget review at a time, and caution is the natural enemy of the kind of tinkering that produced most of the breakthroughs digital leaders now depend on.

Token economics did not create this dynamic out of nowhere. Cloud computing metered usage for over a decade before generative AI arrived. What is different now is the granularity and the speed at which costs accumulate, and the fact that the resource being metered, intelligence itself, sits closer to the center of the business than compute or storage ever did. When the thing you are rationing is thinking, the stakes of that rationing change.

The Freedom That Matters More

All of which leads to a more important question than whether a given model or platform is free to use. The better question is whether an organization remains free to choose.

Look around the current AI landscape and you will find a familiar set of tensions: Proprietary models with closed weights, cloud platforms that make switching costs punishing by design, licensing terms that shift with little warning, export controls that redraw the map of who can access which capabilities, regional availability that varies by jurisdiction, and API changes that can quietly deprecate a feature a whole product line was built on. None of this is inherently sinister. Companies that spend billions of dollars training frontier models have a legitimate claim to a commercial return on that investment. Governments that restrict the flow of the most powerful AI systems have legitimate national security interests behind those restrictions. Frontier AI is extraordinarily expensive to build and genuinely difficult to secure. Painting any of this as villainy misses the point entirely.

The issue was never motive. The issue is resilience. A digital leader does not need to believe a vendor is acting in bad faith to ask a much more practical question: what happens to my organization if this vendor changes its pricing tomorrow, revises its terms of service, deprecates the API my product depends on, or decides, for reasons entirely its own, that a capability I have built my roadmap around is no longer available in my region or my tier. That is not a hypothetical born of paranoia. It is the kind of scenario that has already played out, in smaller ways, across cloud, SaaS, and now AI.

None of this is an argument against commercial software, which has driven more enterprise innovation over the last forty years than any other single force. It is not an argument that every model needs to be open source either. Some technologies require the kind of capital investment that only a commercial return can justify, and demanding otherwise is its own kind of naivete. The argument is for balance. Every healthy technology ecosystem in modern computing history has combined proprietary innovation with open foundations, and the two have reinforced each other more often than they have competed. Commercial vendors tend to push the frontier fastest because they have the capital and the incentive. Open ecosystems tend to keep power from concentrating too far in any one place, and they keep the underlying standards and interoperability alive long after any single vendor’s product roadmap has moved on.

An Architectural Decision, Not a Philosophical One

For a CIO or a Chief Digital Officer, freedom deserves the same kind of treatment that previous generations gave to redundancy, disaster recovery, and cybersecurity. Nobody builds a data center with a single power source and calls it a strategy. Nobody signs off on a security posture with a single point of failure and calls it done. Freedom, in the same way, is an architectural decision. It shows up in choices about open standards, in whether APIs are portable or proprietary, in whether data can actually leave a platform in a usable format, in multi-cloud and multi-model strategies, and in a general discomfort with unnecessary dependence on any single provider, no matter how good that provider’s product currently is.

Boards are not always asking the right question about this, mostly because the right question is uncomfortable and a little unglamorous. It is worth asking, anyway, on a regular cadence, about every critical vendor relationship: If this vendor disappeared tomorrow, or fundamentally changed its terms, how much of our digital business would remain under our control? For a lot of organizations, the honest answer to that question is less reassuring than anyone would like.

What the Founders Were Actually Fighting For

The founders were not fighting for cheaper goods from the crown. Free as in beer was never the point of 1776. They were fighting for the ability to govern themselves, to make their own decisions about their own future without asking permission from an authority three thousand miles away. That analogy should not be stretched further than it can bear. Nobody is comparing a SaaS contract to the Boston Tea Party. But there is a real and useful lesson underneath it for anyone running technology strategy today. Freedom is not valuable because it is inexpensive. It is valuable because it preserves choice, preserves competition, and preserves an organization’s ability to determine its own path forward instead of having that path determined for it.

As the fireworks go up this Independence Day, it is worth remembering that the technologies that actually transformed enterprise computing were never built on brilliant engineering alone. They were built on openness, on collaboration, and on the belief that innovation flourishes fastest when people are free to build on each other’s work, improve it, and share what they learn. In the age of AI and token economics, free as in beer may become harder to find than it used to be. That may simply be the economics of a new era, and there is nothing wrong with paying for value. But free as in freedom is worth protecting deliberately, not out of nostalgia for an earlier internet, but because it remains one of the most durable competitive advantages an enterprise, or a nation, can possess.