Giving everyone in a country free access to a paid AI tool sounds generous. It also sounds like a very smart way to make a whole population depend on one company’s product before anyone has really agreed on what that dependence should cost.
OpenAI says it struck a first-of-its-kind deal with Malta’s government to give all residents free access to ChatGPT Plus. That’s the core fact. A small Mediterranean nation, a national rollout, and a premium AI subscription suddenly treated like public infrastructure. From what’s been shared publicly, it’s framed as access for everyone, not just students or civil servants.
I don’t think this is automatically “bad.” I do think it’s loaded.
Because “free” is never just free. “Free” is a strategy. And when the strategy runs through a government, it stops being a personal choice and becomes the default setting for society.
Imagine you’re a Maltese student. You’d be foolish not to use it. You’d use it for homework help, language practice, brainstorming, and maybe even tutoring you can’t afford. That’s a real win. It could lift outcomes for kids who don’t have parents with time, money, or education to help them. If you care about equal chances, handing out a powerful tool to everyone is hard to argue with.
Now imagine you’re a small business owner. You can draft emails, translate menus, write job ads, summarize regulations, and plan marketing in minutes. That’s not just convenience. That’s time back. That’s fewer paid hours for admin work you hate. For a tiny shop or a one-person service business, that could be the difference between growing and staying stuck.
So yes, there’s obvious upside. The part that bothers me is how quickly “nice-to-have” becomes “must-have” when it’s built into everyday life.
Once people start relying on one tool for schoolwork, paperwork, job applications, and basic writing, the country effectively standardizes on that tool. The winners are the people who learn to use it well, early. The losers are the people who don’t, or can’t, or simply refuse. And the biggest winner is the company whose product becomes normal, trusted, and habitual at a national level.
That’s where the deal stops feeling like a gift and starts feeling like a beachhead.
If you’re Malta’s government, you might say: we’re doing what governments should do—bringing useful tools to everyone, not only the rich. Fair. But there’s a difference between providing access to a general capability (like internet connectivity) and providing one branded service (one specific AI subscription). The first creates options. The second quietly narrows them.
And then there’s the question of what “all residents” really means in daily life. How do people sign up? What data is collected? What identity checks exist, if any? What happens when something goes wrong—wrong answers, harmful advice, a student accused of cheating, a citizen misled while filling out forms? None of that is spelled out in the post you shared, and that uncertainty matters because the tool isn’t just entertainment. People will use it for decisions.
Picture a person dealing with a legal problem, a health worry, a visa issue, or a benefits form. They ask the AI because it’s fast and “officially available.” If it gives a confident but wrong answer, the damage is real. Not theoretical. And when the government is part of the rollout, people will assume the tool is safer and more reliable than it actually is.
This also shifts power in a way most people won’t notice until later. If ChatGPT Plus becomes the standard helper, then whatever it’s best at becomes what people do more of, and whatever it’s weak at becomes what people quietly stop doing. People may write less on their own. Teachers may assign less writing, or more “AI-proof” tasks that punish honest students. Employers may start expecting polished writing from everyone, because “you have no excuse now.” That sounds like progress until you realize it raises the bar for basic participation in society.
There’s a serious counterview, and I get it: if AI is going to be everywhere anyway, giving broad access now is better than letting only wealthy families and big companies get the advantage. Making it free could be a way to avoid an AI class divide. And if Malta can do this without turning it into surveillance or dependency, it could be a model.
But that “if” is doing a lot of work.
The hardest part is that we don’t know what Malta gave up to get this. Maybe nothing significant. Maybe it’s a limited-time pilot. Maybe it’s tightly privacy-protected. Or maybe it’s the start of governments becoming distribution partners for private AI—turning public systems into pipelines for one company’s tools. Once that pattern sets in, other countries will copy it, and competition will shift from “best product” to “best political deal.”
If this becomes normal, citizens may wake up in a few years realizing they didn’t just get free access to an assistant—they helped choose the assistant for the whole country without a real public argument about tradeoffs.
So here’s the debate I actually want: should a government ever make one private AI service the default tool for an entire population, even if it’s free?