These days, as I use AI and find myself hoping for a better model every day, I keep having this thought.

“If it becomes this easy to make things, what kind of future is coming?”

It writes quickly, creates images, writes code, and summarizes things. Work that used to take hours sometimes ends in minutes. On the surface, productivity clearly goes up. The amount of output one person can create also increases.

Then will this increase in productivity bring us more breathing room?

At first, it seems like it will. If we can do more work in less time, it feels like we will have time left over and costs will go down. But things may not necessarily flow that way.

If we become able to make more of something, what happens to the value of each individual result?

It will probably be more likely to fall. In the past, being able to write, design, or code was itself a certain kind of competitiveness. But if AI lets anyone create results at a certain level, the fact that you “can make it” will no longer be special by itself.

Then is the answer high quality?

Of course, top-level results will still have value. But looking at the market as a whole, it is not always only the highest quality that survives. In most industries, “good enough” results are used far more often. There are many cases where what is needed is not a perfect piece of writing but writing that can be used right away, not the best design but a design that can be tested quickly, not perfect code but code that works for now.

In the end, as AI advances, the world may become flooded with reasonably good results.

The problem comes after that.

When there are more results, the price per item can fall. But that does not necessarily mean producers become more comfortable. If the money left per item decreases, they have to make more to earn the same revenue. If they make more, the market gets even more output, and the value of each individual piece falls again.

Put a little simply, the structure looks like this.

“Because we can make it cheaply, we have to make more.”

“After making more, the value per item falls even further.”

“So we have to make even more.”

AI lowers the cost of making one thing, but I am not yet sure whether it lowers the total cost of surviving in competition.

And this change will not necessarily be limited to digital outputs like writing, images, and code. Even before robots reach the stage of directly making or moving physical objects in the real world, AI is already entering our decision-making process. When we wonder what to buy, how to study, what to write, or which choice is better, we are asking AI questions and getting answers more and more often.

Being able to ask instantly and get an answer instantly does not just mean we have gained a convenient search tool. It means more choices and judgments are being made through AI. In that case, AI usage may keep increasing not only in the production processes of certain industries, but also inside individuals’ everyday decisions.

This cost is not only human time. If we become a society that uses AI more, the burden on physical infrastructure such as electricity, servers, GPUs, data centers, and communication networks may also grow. Digital outputs look light, but behind them is fairly heavy infrastructure. One piece of content may become cheap to produce, but the cost of the production system run by society as a whole may move in a different direction.

This is where a slightly uneasy thought comes in. If we end up using more resources in order to produce more, the effect may not end inside the AI industry. Resources like electricity, semiconductors, servers, and communication networks are foundational resources used by other industries too. If competition in AI production pulls more of these resources in, won’t that create pressure on the costs of other industries as well? If so, are we becoming more abundant, or are we instead living on top of more expensive infrastructure?

The game industry gives me a similar feeling.

Games no longer compete only with other games. Even without playing a game directly, users can get plenty of entertainment from YouTube, streaming, shorts, and communities. They can just watch strategy videos, streamer broadcasts, or highlight clips.

Then it is not enough for a game company to simply make a good game. To win people’s time, it has to care about videos, communities, updates, events, influencers, and buzz as well. In the end, the cost of creating satisfaction does not arise only inside the game.

Costs can also rise on the gamer’s side. For a better experience, people need better graphics cards, consoles, monitors, storage devices, and network environments. The price of a single game is not the only issue; the surrounding costs required to fully enjoy that game rise together. In the end, a structure may form where both producers and consumers have to bear higher costs to get the same satisfaction.

The job market may be similar.

If AI polishes resumes, cover letters, portfolio descriptions, and interview answers well, job seekers can apply more easily. But if everyone can submit convincing documents, it becomes harder for companies to judge people from documents alone.

Then companies may demand more assignments, interviews, tests, and reference checks. Job seekers apply more but receive fewer responses, while employers receive more applications but find it harder to feel confident.

In the end, both sides may end up moving more, while satisfaction goes down.

Seen this way, the changes AI brings probably will not end simply with “things become cheaper and more convenient.” Some things will certainly become cheaper. Writing, image creation, translation, and summarization may become much easier.

But at the same time, other costs may rise.

The cost of having to make more.

The cost of having to verify more.

The cost of having to compare more.

The cost of having to prove more.

The cost of having to keep up so as not to fall behind.

Even so, on the other hand, it also seems clear that an era has arrived in which it is easier for individuals to build their abilities. Things that previously required asking someone, taking lectures, or looking through books can now be started much more lightly. There has never been an era when a tutor that gives constant feedback, accepts questions, and explains again the parts I do not understand has become this inexpensive.

So sometimes I also think the opposite. If everyone can learn, make, and experiment much more easily than before, won’t individual ability rise significantly overall? If everyone comes to have a small teacher, a small adviser, and a small work partner next to them, won’t we be entering an age where everyone soon becomes something like a superhuman?

But because of that very possibility, the question returns to the beginning. If everyone can learn more, make more, and grow more, will that give us breathing room? Or will it bring higher standards and fiercer competition?

If AI raises productivity, will we be able to enjoy it?