Right now everyone wants to earn more with the help of artificial intelligence, and it seems that this is the ideal moment for it. What used to take hours, or needed a whole team, we now see one person doing many times faster. But the paradox is that earning has not got easier. Clients are still just as hard to find. Especially hard. Competition is growing very strongly, and the demands on a specialist only rise. So where is the money in artificial intelligence really today?
And what separates those whose income artificial intelligence has already increased from those who have simply started working faster? That is what we will talk about today. Hello everyone! You are on the ToTheMoon channel with Alexander Volchek. Today is our special episode, the one that comes out on Wednesday, Friday, sometimes Tuesday, with other people. On Sunday we have our shared joint podcast. Yes, we already have more than a hundred and fifty episodes about artificial intelligence and technology from Silicon Valley and around the world.
So, what did I want to discuss with you today? It is the question people ask endlessly about artificial intelligence: how do you make money in artificial intelligence or with the help of artificial intelligence? There are the stories, the series, that we filmed from the point of view of professions. I was discussing exactly this recently: what will happen to professions, how professions will disappear or not disappear, appear or not appear, what opportunities there are. Right now I want to film where the money actually is, where the earnings are and how to approach this.
I want to say straight away that to approach the question of earning, we are not going to look at it casually — that there is some tool, some piece of software, and it just produces money. We will look more broadly at the whole chain of earning. Is that earning possible at all? Take, for example, the fact that with artificial intelligence you can now process and create texts very fast. And clearly there is still an enormous number of orders in the world on this subject, and you can sell your time reasonably well in the part that concerns, say, creating posts on social networks or writing articles for a site, or some other editorial work.
But at the same time, what was and still is the main problem that existed before artificial intelligence?
It is how to find yourself a client. The problem is that even if today we see an enormous number of opportunities in artificial intelligence that let you speed up — quickly produce analytical reports, quickly gather some information, some data, in other words actually do a great deal of work fast. And obviously in many fields artificial intelligence has sped this up somewhere by a factor of a thousand. Or made it possible at all. Then on the other side, how do you find that client?
For me, for instance, as a client, I have a great deal, a great deal of demand across different tasks. I can say that the ease of finding contractors for various work has not increased for me, that is how I would put it. With the arrival of artificial intelligence. That is, over the past three years I personally have not felt that it became easier for me to find people. It would seem that the number of people who can now do certain work — marketing, sales, analytics, or, I don't know, editorial work, or writing code — that number of people should have grown. But I do not see a correlation with people becoming easy to hire.
And it is not just me who does not see it. I am in a wide circle of contact with different businesses, I sit on many boards or own various companies. And I do not see it in those other companies either, or in talking with different people, that anyone says so on a mass scale. Clearly it depends on the industry. That it has become easy for me to find people — because you see what the point is: on the one hand you, as a person who wants to earn money with artificial intelligence or technology, you want to earn more money.
On the other side, where companies commission people, they want to pay less money, or to get better effectiveness. Let us put it this way: not everyone wants to pay less money. There are projects where they are ready to pay the same but to increase, say, effectiveness or find people more easily. That is, if my risk in hiring a person on some project is high, then how do I find a person so that I do not have that risk? What has happened overall is that I have not noticed the correlation. Tell me, have you noticed the correlation in your projects?
If any business owners are here — and I think there are people here who hire someone. I have not noticed it. It has not got easier. Although, once again, there should have been far more people, because artificial intelligence — everyone talks about this, look, absolutely all company owners say it. Just recently Sam Altman said that he believes modern technology will create an additional, simpler opportunity for ordinary people to build businesses, for ordinary people to find work more easily, for ordinary people, for more and more people, to realise themselves well and properly.
And you get an interesting circumstance. That it looks as though this should have happened, but if you look from the employers' side, you do not see that volume of people. What is more, I want to say that the people I know, and I myself, can say: my requirements have somewhere risen — or rather it is not that they rose, it is that my approach changed as to whom I want to hire. Which means it has, if anything, become harder for me to hire than before. That is, it did not get easier, it got harder. You want a person with a different spread of professional qualities now.
So if you want to earn money with technology or with artificial intelligence, or to increase your earnings somehow, it is important to understand that problem number one is, of course, finding a client. And anyone who says that with artificial intelligence there are now an enormous number of ways to find clients — that is an abstraction. That is, if you are an artist and you are now going to start opening your own business and looking for clients, and learning marketing and sales, with artificial intelligence supposedly helping in that — you have to understand that it has definitely not got easier for you.
In effect it has got harder for you now, because artificial intelligence helps, first of all, not newcomers; it helps those who were already present in the market.
And it will help them far more. That is how it is. Clearly, if you have broad thinking, and marketing or sales turned out easier for you — well, fine, you gained some advantage, but the question is: did they turn out easier? Because in the world today, from the marketing side, it has become even harder. Once again: the greater the competition, the harder it is. The more people appear who understand something, the harder it becomes to sell. This is by no means a position from which you should tell yourself: then I have
no chance at all, it seems, as a person, when it comes to earning. What I want to show here is rather the way you can, in your own life, see additional opportunities, in particular with artificial intelligence. And the additional opportunity I see today, number one, is based on you needing to know very well, and use heavily, within your basic life and the processes that already happen for you every day, ordinary chats such as ChatGPT or Claude. What does using them mean? It means that in a day you should be having, through these chats, no fewer than thirty-five, fifty, or seventy different requests. That is, these systems have to become a simple, ordinary norm for you, an ordinary norm, so that you merge with them.
It is like how you once had to know Excel, some basic simple functions, or to know Microsoft Word, the ordinary functions; now the task is to know the basics in ChatGPT, in Claude. Again, for someone this may sound banal, and someone will say this is all some sort of nonsense. But if you look at the facts of what is happening — we record the Sunday podcast, and a very good example in this case is Tanya, who has not started using, say, Codex
because she does not bring Codex in — not ChatGPT, but she does not bring Codex into her simple projects. And the next step, which follows on quite closely, just more deeply. If you genuinely want to earn with artificial intelligence or with technology, it is, of course, to start applying ChatGPT, Codex and Claude Code in your work. What does applying mean? It means that probably fifty to seventy percent of the various tasks you do, you start doing not in a plain chat — Claude, or ChatGPT, or Gemini — but doing it yourself with Codex.
And do it — look — using not third-party software. There is a great deal of third-party software that lets you do something easily. For example, using some speech-recognition system, rather than doing it inside Codex. Or using a separate tool to create good PDF decks or PowerPoint decks, in Figma, say, or in Miro, rather than doing it yourself inside Codex or rather than doing it inside Claude Code. So the task in front of you is to reduce the number of third-party technology tools, especially the ones involving artificial intelligence, and to start doing more business processes directly yourself in systems like Codex or Claude Code. Why do I point to these two specific systems?
Because as of today they have a fundamental difference from everything else. A fundamental difference. That is, they differ from Gemini, they differ from Grok, they differ from DeepSeek. Those systems do not have such capabilities. I am talking right now about Claude Code or about Codex not because I have got fixated on these two
systems. Sometimes it seems to someone that someone has got fixated on something. It is because you have to use the best in your life, not make do with leftovers somewhere. Because you may have restrictions within your company on using different models. You may not have access to certain models because of your location. But in general, separately — but in general you personally, for personal use, can connect to practically most of the best systems. And get access to them. Clearly, where it is permitted by law, by the possibilities of the systems themselves and so on. Here, of course, I fundamentally recommend having some minimal paid tier.
These conversations about how every person can now open a business for free, I accept them relatively, because if you cannot pay twenty dollars for some entry-level paid tier of Codex or Claude — someone will say that means I do not live in that world, that for many people twenty dollars is expensive. Then do not have the conversation about wanting to earn five hundred dollars more, or three hundred, or seven hundred. Think about it this way. Suppose you earn fifteen hundred dollars a month. Can you put in an extra hundred dollars in order to raise your salary by four hundred?
People fundamentally think very strangely. They say: first we will raise our salary by five hundred dollars, then we will think about whether to pay twenty dollars. And that is exactly the thinking behind why people still do not buy the two-hundred-dollar pro subscriptions. But if you genuinely want to increase your earnings, well, you have to make certain investments. And what is more, these systems where you pay twenty dollars, and they give you an effect — that is two hundred and forty dollars a year, or, three hundred a year, taking a whole year. Over a year, even in terms of your basic life, your health, the effectiveness of dealing with other people, you will very likely make that money back. If you, well, cannot pay, there is a free version.
Clearly you can use the free versions, including Codex and including Claude Code, which are available free. Yes, on a certain number of tokens, but they are available free in any case. You can take some actions and find your way around. Those actions in Codex or in Claude Code should not only be the ones to do with work; they should touch your life. As in the examples I give, where I connected something, sorted out some of my finances, or connected a Tamgee camera, or set up some family Telegram notification, or some simple things, ordinary things that belong to ordinary life, or even sorted out files on a drive, or some photographs on a drive.
Those kinds of things, those matters, also need doing in these systems. Do you need to study and take courses somewhere on how to work in Claude Code
or how to work in Codex? You do not! Because if you use chats through the day, making thirty, forty, fifty different requests a day on the various subjects of your life — from your health, your condition, I don't know, flowers, plates, dishes, food, through to any work questions, your learning and everything else — then you will learn far faster and better to use all these systems. Far better. What is more, in any of these systems you can always ask how to work in that system.
I want to say straight away that cheap, inexpensive or free systems can differ quite seriously in the quality of their answers when it comes to new functions. Believe it or not, but the new releases and new technology, and the most modern artificial intelligence, is available only in the top subscriptions and the very latest models. In any case. And once you have got access to that, you have the chance to learn to work with it faster. By the way, I want to say straight away that in the current reality of using artificial intelligence, in effect, you have no way of preparing for something in advance.
Just as I have no way of preparing for something in advance. I mean learning, mastering how to do one particular business process in Codex or in Claude. What a year ago you had to be a very clever person to do in Codex or in Claude. Now many processes have become absolutely elementary. Tasks are easily done. What is more, we are moving into a time when the effectiveness of work done will fly ahead. And therefore the task is not to learn to do some one simple action; the task is to understand how artificial intelligence works in general, and to be able calmly to do it and not be afraid to do it. So, once again: I am not against these training programmes, but I would look at any programme very clearly from the point of view of the investment. If that training programme costs a thousand dollars a year, I would rather suggest you spend a thousand dollars on the hundred-dollar pro subscription to ChatGPT.
The effectiveness you get will be several times better, usually several times. Again, not necessarily, not guaranteed. There are different systems or different cross-sections of what may happen. Since the number of professions and the number of lives, and different ages, and men and women, and different regions overall
differ enormously, every person of course has their own circumstances as to what they need to do in order to increase their effectiveness. That has a very strong influence. And, probably, I personally observe how a large number of people are coasting right now, using artificial intelligence in their work, selling their services just as before. Although the conclusions they draw, in terms of quality, are of an altogether different kind. I had one interesting meeting today, and one person said at that meeting that — he says: can we hand our analysts the information we have just been discussing?
Because the way you produced it. Let our analysts prepare for meetings like that. But it turns out that I prepared for the meeting not because I am better than those people professionally, I don't know, in technology analytics, but because I have this kind of thinking system, and I drew those conclusions. And however much you show those people my conclusions, they will not be able to prepare the same way. That is, every person, if you look at it properly, absolutely every person will still be in their own element, will still be in their own environment in terms of what they do, how they do it, where they do it. And only you yourself can decide how to improve your effectiveness and where not to lose out.
That possibility, of course, always exists. And our main task, for instance, my main task on the ToTheMoon channel — given, by the way, the circumstances of my life, I do a great deal of personal consulting, meetings, where I help people and companies, of course, to make money. I can say that the only advice that can genuinely be given in answering this question, on how in this subject today, from the point of view of artificial intelligence, is to use that artificial intelligence in your own work.
Only then can you become a genuine, true professional at it. Which is what I wish for all of you. And here I have one small addition. Those of you who watched to the end, please write in your cases and questions. And what do cases and questions mean? Tell me your examples. I will try to work through them and answer some. Because when you send real, genuine, working situations, I have the chance to give examples: if the situation is like this, then perhaps this approach is better; if the situation is like that — that approach is better, because otherwise it is a certain abstraction. And someone in the comments may discuss, I don't know, someone who really liked my appearance, or someone who liked, or did not like, how many words a minute I said; someone may concentrate on getting effectiveness for themselves, an additional opportunity in terms of their own life. There you go, and you can do that too.