Over the past few weeks, I've spent a significant amount of time promoting my AI course and seeking affiliates to help sell it. In this process, I reached out to several social media influencers who have a large following and often post content about AI. My primary focus was LinkedIn and Twitter, where AI discussions are thriving.
To my surprise, almost every one of them declined to promote my AI course, even when I offered them a generous 75% affiliate commission. Instead, they were willing to talk about my course if I paid them upfront. This didn't make sense — if they were truly genuine in promoting AI and its potential, they should have been more interested in the affiliate commission that could lead to bigger long-term earnings. With their huge followers interested in AI, they could easily earn a lot of money from my affiliate offer compared to a fixed upfront price. But as they prefer a fixed payment to an affiliate offer, they know that their huge following is not real; it is inflated.
This situation got me thinking, and I decided to dig deeper into these so-called AI influencers. What I found was unsettling: a clear pattern started to emerge. It seems that some AI tool makers are funding these influencers to get inflated/unreal/low-quality huge followings, pushing tools that are more about hype than substance, with the help of an inflated fanbase. With this process, they are trying to create hype about their fancy but useless AI Tools. I fear this kind of hype could destroy the entire AI ecosystem, much like what happened with Web 3.0.
We saw how overhyped promises and so-called “experts” in the Web 3.0 space made it difficult for genuine projects to get the attention they deserved. Today, we need to be cautious and mindful as the same trends threaten to take over AI.
This realization led me to write this post. It’s important for us to recognize the difference between genuine AI progress and the noise created by hype — and to handle this technology responsibly.
We Are Living in the Era of the AI Echo Chamber
We are all hearing about Artificial Intelligence these days. Whether it’s ChatGPT, Google Gemini, or Copilot, AI tools have suddenly become part of everyday talk. Everyone wants to use AI, sell AI, or look like they know AI. But here’s the problem — most people are only scratching the surface. We are not really in the age of intelligent machines; we are in the age of people pretending to understand them.
Everyone’s Talking, But Few Are Thinking
Today, AI is a hot topic. It’s everywhere — news, social media, WhatsApp forwards, online courses. People are throwing around words like “prompt engineering,” “AI expert,” "Agentic AI", "Vibe Coding", and “machine learning” without knowing what they actually mean. It’s like a classroom where everyone is talking, but nobody is really learning.
People are creating videos, courses, and posts about AI after spending just a few hours with ChatGPT. Many influencers are giving advice with full confidence, even though they have no background in AI or computer science. Companies are planning "AI transformation" without even asking their technical teams how realistic it is.
Echo Chamber of Confusion
This creates what we call an "echo chamber." People repeat what others say without questioning or understanding it. On LinkedIn and other platforms, the same ideas are repeated again and again — just with different packaging. If someone questions the hype or tries to explain the limitations of AI, they are often ignored or labeled as negative.
Why This Is Dangerous
This fake understanding is not just a small issue. It can lead to serious problems:
Companies wasting money on AI projects that don’t work
Poor decisions made based on wrong AI results
Biases in hiring, medical treatment, and loans because of faulty algorithms
Overtrust in machines that don’t really understand anything
In a country like India, where millions are still learning about digital tools, this misinformation can spread fast and cause big mistakes. Imagine a school using AI to grade students or a startup using AI to hire people without checking if the model is fair or correct. And almost all big companies are offering their AI Tools freely initially. But people will start realizing the limitations in the free version only after they become heavily dependent on those tools. Then they won't have any option other than going with the paid version. It reminds me of how Microsoft Windows dominates many countries like India, even when the free and open-source operating system Linux, is better than Windows.
AI Is Not Magic — It’s Just Maths and Code
Let’s be clear: AI is not a human brain. It doesn’t have feelings. It doesn’t “think” or “understand” the way we do. AI is a tool — a powerful one — built using data, probability, logic, and mathematics. It works based on patterns it finds in the data we give it. If the data is wrong or biased, the results will also be wrong.
Many people feel AI is some kind of magic. But the truth is: it’s just well-designed software. The moment we treat it like a superhuman, we lose our ability to control and question it.
What Should We Do Instead?
Instead of blindly following the hype, we should:
Learn the basics of how AI works (not just how to use tools)
Encourage open discussions, not just one-sided promotions
Work together — developers, designers, educators, policy makers — to use AI responsibly
Stay humble — the more we learn, the more we realise how little we know
You don’t need to become an AI scientist. But if you are using AI in your work, business, or content, it’s your responsibility to understand it properly. Even simple courses or books can give you a strong foundation.
We Are at a Crossroads
Right now, we have two choices:
Continue spreading half-knowledge and build a world of confusion.
Slow down, learn properly, and build a future where AI helps society in a safe and useful way.
The first option is popular, quick, and gives short-term fame. The second one takes time, effort, and honesty. But in the long run, the second path will create better jobs, better tools, and a better society.
Let’s choose wisely.
There is no question about the huge development of AI, the opportunities it is going to give, and the impact on jobs. Surely we need to learn AI. There is no question about it. That's why I created my AI Course to share my understanding about AI. I am criticising only the unnecessary hype.
And when learning AI, you need to be very careful. I used to get questions like "why do I need to buy your AI course while many free AI courses are offered by big companies?". But if you analyze those free courses, you can understand the motive behind them. Mostly, they will be preparing the learner to use their own AI Tools or cloud services; they won't tell about better open-source and on-premise opportunities. So, it is important to learn AI from an unbiased source.
Use the Coupon code QPT to get a discount on my AI Course available at Rajamanickam.Com
Updates On 3 May 2025:
One person made below comment in a Reddit discussion about this post.
"To my surprise, almost every one of them declined to promote my AI course, even when I offered them a generous 75% affiliate commission. Instead, they were willing to talk about my course if I paid them upfront. This didn't make sense — if they were truly genuine in promoting AI and its potential, they should have been more interested in the affiliate commission that could lead to bigger long-term earnings."
The author draws their conclusions from the fact that advertisers and influencers would rather get paid directly, rather than risk not getting paid by relying on affiliate link commissions for the courses the author is flogging. This comes across as resentfulness, rather than informed conjecture. Could it be that the author's courses might not sell well?
That issue alone is enough to dismiss this article, but reading on, I'm not sure what this piece adds to the conversation, other than "some people have an inflated view of the state of AI right now" and "we should all try to know what it is we're talking about when we talk about it."
This is, essentially, a badly written advertisement.
I posted the below reply to him.
Let me explain this in detail. Assume that you are an Influencer with 100K followers/connections. You post daily content about AI, and many people like and share it. In this situation, I approach you with my affiliate offer. You decline it and instead provide me with an offer to post about my course (without an affiliate link) if I pay you $100. In case of an affiliate offer, you can easily earn more than $1500 with a 100K highly relevant audience and a 75% affiliate offer. Why should you choose just $100? It is a big question. From my side, theoretically it is a good opportunity. Just by paying $100 upfront, I can save $1400 later. But the question why you chose $100 over $1500 stopped me from taking your offer and made me think about the reason behind it. That's how I started my analysis and came to the conclusion about inflated followers. My assumption is like below. It may not be exactly like this, but something similar to one.
A greedy person develops a useless AI Tool and projects it in a fancy way. Now no one will believe him, and he won't earn money from his useless AI Tool
Then he spent some money to find some influencers in AI and form a network of them. In a coordinated way, they increase the connections/followers to make them look like a great influencer in AI.
Then all of them will start sharing AI-related information and mutually promote them as a network. Many people will believe them and follow them.
Once after gained reputation and a huge following, they will start implementing their original agenda of selling the fancy, useless AI Tool. Even if it is useful, there will be some traps, like huge subscription fees in later days once the users are locked in with them. Now they will get back all the money they spent and earn more than that.
It is their business. I don't have the right to ask them. But overall, this kind of AI hype spoils the entire AI ecosystem. Because those who spend money on those fancy AI tools will start complaining about not just that tool, but the entire AI system. And, this kind of forced marketing will bury the good AI tools, too.
I hope you understand what I am trying to explain here. So, it is our responsibility to create awareness about it.
And it is nothing related to my course. I am not saying that all the people declining my affiliate offer are like this. I understand that my course is not the best one, and I am not a well-known person. It is difficult for the affiliates to sell it. So, they are declining the offer. I 100% accept them. A lot of small-scale honest affiliates also declined my offer. I understand their thoughts. I am not blaming them. You should understand the key difference. These honest affiliates are not giving an offer to promote my course for a fixed price. They are saying that I don't have the audience to sell the course. I appreciate their honesty. I hope you understand this difference.
Updates On 4th May 2025:
I got below reply.
You are asking influencers to advertise your product for a commission, rather than for direct payment. This strategy will not be successful if influencers are not convinced your courses will sell well. If no one signs up for your course, the influencer gets nothing in return for cheapening their brand.
I gave the below reply.
I understand your explanation. As a YouTuber, I used to do the same thing, i-e, rejecting affiliate offers and expecting a fixed price normally. But if I get a good affiliate offer, I won't reject it for a low fixed price. For example, if you offer me a 75% affiliate commission for a good AI-related Gumroad product, I will add the link immediately to my sales page. My key point is not about it. I just explained these things to explain why I started doing more analysis about the influencers.
No comments:
Post a Comment