Subject of the Review

The author of the review runs a YouTube channel. According to him, over the past two months the channel brought in 1.2 million rubles. To back this up, he shows statistics and several of his own channels on American YouTube:

• one channel brought in several tens of thousands of dollars;
• another brought in 788 thousand rubles;
• a third brought in 125 thousand rubles;
• a fourth brought in $2,100, gained 160 thousand subscribers, and received YouTube's Silver Play Button.

The episode is devoted to videos and courses that promise income from AI-generated content on YouTube. In the author's view, these materials distort the statistics, conceal risks, and create financial dependence on paid services.

Verifying the Claimed Results

The subject of this review is the video "How to Earn from Zero to $8,655 a Month with a YouTube Channel. Step-by-Step Guide." Its author, Deimin, claims:

• he has been creating content for more than 3 years;
• he has created more than 20 successful channels;
• he earns $5,000 a month;
• he shows the results of his students.

Channel Sample

Of the 20 channels claimed, Deimin showed four. The reviewer believes the other 16 did not bring in significant income. By his estimate, such channels earned about $100 over six months.

Of the four channels shown, only two demonstrate stable dynamics:

The author notes that Shorts monetize significantly worse than long videos: with four times as many views, the income was almost five times lower.

The other two channels showed the following picture:

• The middle channel brought in $2,700. Then no income came in over the following six months.
• The bottom left channel was hit by YouTube's algorithmic restrictions. It has generated no income since January 2025. The last date in the statistics is June 7.

The author considers Shorts a "pure lottery": the channel constantly risks being blocked and losing recommendations.

Example of a dinosaur channel

The author analyzes a channel with videos about dinosaurs and other dangerous animals, which, according to vidIQ, is gaining steady views. Figures for the year:

• 7 million views in total;
• June: 133,000 views;
• July: 1 million views;
• August: 2 million views.

The author assumes an RPM (revenue per 1,000 views) of about $5. He explains this estimate by the short running time of the videos. On this basis, he calculates revenue of about$5,000 in July and about $10,000 in August, that is,$5,000 to $10,000 per month. This is his calculation, not YouTube data.

Лотерея просмотров

Автор описывает распределение просмотров как случайное:

• примерно 1% роликов набирают миллион просмотров;
• 99% роликов набирают 1 000–2 000 просмотров;
• отдельные ролики выходят на 300 000–400 000 просмотров даже у самых успешных AI-каналов, но стабильных каналов с высокими показателями нет;
• разброс у AI-каналов составляет от 500 до нескольких тысяч просмотров.

Автор приводит примеры резких колебаний:

• ролик набрал 814 000 просмотров, следующий — 13 000;
• ролик набрал 2 000 просмотров, на следующий день — 392 000;
• канал о быстром питании имел колебания от 480 000 до 2 000 просмотров.

Автор сравнивает такие каналы с лотереей и подчёркивает, что успех одного ролика не воспроизводится. Он также указывает, что пользователи на Reddit годами не достигают порога, при котором хотя бы одно видео выстреливает.

Example of Checking a Copycat Video

The author checks whether a successful video can be reproduced with AI. He describes the procedure:

1. A link to the original video is uploaded to a neural network.
2. Prompt: analyze the video, describe its content, and prepare a video in a similar style with possible changes to the script.
3. A duration of 2 minutes is set.
4. The neural network returns a script, for example, beginning with a morning alarm clock ringing.
5. Result: the video is nearly identical to the original.

The author notes that a channel with videos about ancient people has 15 million views and stable performance (280,000, 269,000, and 288,000 views on individual videos). In his assessment, this channel has a creative concept, and AI is used for animation. However, the author considers most similar channels to be formulaic.

YouTube Policy: Changes Effective July 15, 2025

The author cites a YouTube help article. According to it, the policy on repetitious content has been renamed the inauthentic content policy. It now covers videos made on the same idea, with the same animation, and with minor changes.

Key provisions:

• "If you earn money on YouTube, your content must be original and authentic."
• If content is borrowed, it must be substantially modified.
• "There should be no mass production. Artificial intelligence is about mass-producing templated, repetitive, or manipulative content."
• Content must be created "for the enjoyment or education of viewers, and not solely to attract views."

The author explains that the request "make something similar" falls under these criteria. YouTube does not monetize such videos.

According to the author, moderation checks channels when they attempt to enable monetization. Signs of mass production include regularly publishing templated videos, for example daily.

Channel Network Experiment

The author analyzes the results of channel networks created using the infobusiness model.

General Indicators

The study group included 56 channels:

• revenue: 3,400 (the unit of measurement is not included in the analysis);
• views: 2–3 million;
• average RPM: 1.47;
• subscriber growth: 12.5 thousand.

Historical Network

• Peak figures: 35,000 views per day.
• The decline began after monetization was enabled and did not recover.
• One channel in the network received 93,000 views, and the next one about 10,000. The other videos gathered about 1,000 views each.
• One video accounted for 56% of the channel's traffic. According to the author, this is what destroys channels, even those with exclusive content.

Philosophy and Space Networks

• Philosophy network: peak of 22,000 views per day.
• Space network: peak of 23,000 views per day. This figure was provided mainly by one channel.
• The space network has six channels. None is monetized, although the channels have been run for about four months. One of the five channels reached the monetization threshold but did not receive monetization.
• One channel in the space network brought in about 20,000 views per day, which the author considers a high figure for an automated channel. The content is created by AI. Monetization was not granted for it either.

Women's Network

• Nine channels.
• Two channels reached the monetization threshold but did not receive monetization.

September

According to the author, in September not a single automated channel received monetization. The author notes that screenshots of monetization approvals were published in previous weeks, but there were no such publications in September.

Manual Channels

The author notes that the manual channels run by his partner Pasha are growing. However, the manual format takes time: 10, 20, or 100 channels are needed for stable results. One network had 66 channels. The author believes Pasha is the main creator of these channels, while he himself received about 3,000 dollars. He emphasizes that Pasha's results are not his own.

Monetization Forecast

The author expects that 30–50% of channels will receive monetization if they reach the threshold. When running 10 channels, 3–5 at most will receive monetization. At the same time, the likelihood of being caught by filters for automated content is higher.

Causes of the Decline: Mechanics and Statistics

The Algorithm Change on September 14

The author claims that around September 14, YouTube sharply reduced the display of older videos. Top videos stopped getting impressions. Channels that previously generated income saw their earnings drop by 5–10 times. According to him, this leveled all channels: none can carry the income of the others.

The Law of Large Numbers

The author compares the situation to a casino. Early results can be winning, but over the long run, the mathematical expectation becomes apparent. He links this to the law of large numbers and notes that a detailed breakdown with formulas is published in a separate video.

The Dopamine Hook

The author describes the mechanism that keeps creators engaged. In his view, YouTube gives a new channel one video with many views, for example, the tenth video with 20,000–30,000 views. The author expects this result to repeat, but subsequent videos get fewer views. He compares this to winnings in gambling, which create confidence that the result will be repeated.

Income from a Single Video

According to the author, most channels rely on one or two videos for their income. In one case, a channel reached monetization only thanks to a single video. The other videos did not pay back the effort of creating the channel.

The Value of Creativity

The author believes that automated content has no value for YouTube. As a result, creative work turns into a matter of chance. In his view, YouTube originally rewarded creativity with views and money, and automation removes this factor.

Losses of Time and Money

The author points out that, in addition to time, creators pay $100–200 per month for AI services. Manual work, by his estimate, demands more time than working as a courier for Yandex and brings in less money.

Frame Loop Statistics

The author analyzes statistics from Frame Loop, a company that develops AI-powered video tools. According to these statistics:

• faceless AI channels on YouTube earn from $3,000 to$15,000 per month;
• only 3% of channels earn this kind of income;
• 97% of channels do not pay off.

The author notes that these figures reflect the impact of YouTube's 2025 policy update, which restricted monetization for such channels. A significant share of faceless channels fail YouTube's approval process even when they reach the required view counts.

Service Tiers

According to the presentation, the toolset costs about $80 per month. The author points out that the presentation states a payout of about$3 per 100 views, while in his own calculations he uses $3 per 1,000 views.

Calculating the Break-Even Point

The author presents a calculation for a target of $3,000 per month:

• RPM for long-form videos without a face is about $3 per 1,000 views; the range is from$1 to $4;
• to earn $3,000, 1 million monetized views per month are needed;
• with 40 videos per month, each video accounts for about 25,000 views.

YouTube monetization requirements:

• 1,000 subscribers;
• 4,000 hours of watch time.

The author identifies two problems:

1. Until monetization is reached, there is no income. Service payments continue for months or years.
2. Only individual videos that become popular reach around 25,000 views each month.

Short videos do not solve the problem. For a million views, Shorts bring in $50–500 depending on the niche. The author concludes that a viral video produces a spike, not a steady income.

He puts the conclusion this way: the problem is not the $80 per month, but that amount multiplied by six months without payment.

Role of Tools and Automation

The author argues that the choice of tool does not determine the result. Comparing ElevenLabs, ChatGPT, Claude, and other services, he says, does not change the situation. In his view, only a unique point of view and an editorial reason for the video to exist are rewarded. This, in his assessment, is what gurus do not sell in their courses.

The author notes that operators who use AI to reduce the cost of a single video to $1–5 earn more than$10,000. In his estimate, if a human did this work, the cost would be 10–20 times higher. Automation, in the author's view, is a tool for increasing profit, not a product.

The author adds that the results infobusiness marketers present as personal often represent a team's income. For example, an annual income of $50,000 may be split among five people.

Forecast and Conclusions of the Author

The author predicts:

• by 2027, connecting monetization will be extremely difficult;
• YouTube's filters and the monetization threshold will tighten due to the growing volume of this kind of content;
• within just a few months, no one who starts now will get monetization.

In his opinion, info-business promoters are repackaging the scheme by showing fresh success stories of students. Such students, according to the author, reached monetization through unique content and personal time, not by following an instruction. The author compares the promise of an "automated farm" to handing out a pickaxe for manual coal mining.

The author describes the typical path of a scheme participant:

1. Monetization does not get connected.
2. Channels do not gain views, and the reasons remain unclear.
3. The participant buys paid materials and tools.
4. About 90% of participants earn nothing.
5. About 10% receive small amounts, which quickly decline, for example from $1,000 after three months to$100.

The author claims that the success stories are formed from a sample of 100–300 people: the results of ten are published, while subsequent earnings are not shown. In his assessment, the scheme is based on statistical manipulation and exploitation of cognitive biases, rather than on the operation of YouTube's algorithms.

Summary

According to the author of the review, schemes for earning money from AI videos on YouTube:

• rely on a selection of successful examples and do not disclose the results of most participants;
• require ongoing payments for services while producing no income before monetization;
• fall under YouTube restrictions that have been in effect since July 15, 2025;
• carry a high risk of channel blocking.

The author believes that the only ones who actually earn money from such schemes are the organizers who sell the methodology.