This post is a response to a long standing itch. It is a stimulus which can be expressed as a light tickling sensation, just in a non-pleasant way. And like all tickling sensations and itches, the more I scratched it, the more painful it became. This post is long. If you want a short version, please read the TLDR at the end of this post.
Somewhere around 2018, I got this book delivered to my home as a physical copy.
Being honest, I did not go through it except for the first few pages. I wanted to learn more though and since I don’t love learning by reading a lot on paper (especially about computers anyway), I started looking into other places. So I got to learn a few basics about how machines are supposed to learn from the internet.
However, I would like to go two steps back, one step at a time.
When I was in my college, I was taking the ISE (Information Science and Engineering) course. This course was almost 90% of what the CSE course was. Towards the end of our courses, in the final two semesters, the subjects diverged. It was during one of those semesters that I came across the term “Neural Networks”. It was a subject that ISE people did not have but CSE had an option to learn it. Before that, I had only imagined (sometimes) the internet as the kind of network that our brains contained. The first time I got to know about this subject and enquired a little around the topic from my peers and the internet, I realized that it was something that I had once upon a time pondered upon as a child and was really curious about. And that takes me to the previous step.
When I was a child, studying in Class 7th or 8th (not sure when), I was curious about a lot of things. I had barely been introduced to computers (I had seen someone working on MS Word, I guess, and we had a QBASIC class) and I loved history and space science. Of course I wanted to know more about anything and everything. How the supercomputer inside my own skull worked was as intriguing a topic as any other. It was around that time that I had gotten myself a book which talked about how we learn. I think it was this one. Anyways it talked in basics about how things inside my own head worked. Neurons and Dandrite (oh I remember a very powerful glue by that name too) and something called as “Glial Cells”. I have forgotten most of it though. I don’t often complete all the books I buy so this one must be collecting dust in some corner too.
How does it all connect?
So yeah, why am I saying all this? The reason is - I am in a habit of simulating a lot of stuff in my head before I start working on anything related to it. Sometimes, that backfires and at other times, it works like a great lesson, solidifying things in my mind. I think reading that book (in parts) and learning about what neural nets are helped me understand one thing clear - how do we learn anything.
When I started reading that Machine Learning book by some good fellow (pun intended), all those memories and fundamentals that I had kept in my mind got reignited. With the explosive speed at which LLMs took over after ChatGPT arrived, those fundamentals have come back multiple times already in form of YouTube shorts, blog posts, Tweets etc.
The process of learning in items - both with anima and without is similar (well, we modeled both our Gods and our Slaves after us, didn’t we?). You take a lot of data, feed that time and again into a set of neurons, vectorized for the machine (digital numbers for an AI, electrical signals for the brain) and allow the machine to store changes caused by those inputs. That is how a child learns to identify her mother, or toys - because in those first few months, her brain processes millions of images every day and saves the changes caused by that input.
Intelligence
This, more or less, is the basis of intelligence. That right there is the value of practice. It does not matter what you practice - running, cycling, fishing, driving, reading, singing, playing, programming, you name it and the process is the same. That is why “practicing” matters. That’s why “Practice makes you perfect”. The facet that I also want to present is that if something was learned after a lot of data input, which for a living brain in real world would translate to repeated exposure to a certain kind of pattern could create a strong “habit” which is difficult to break. The controversial, yet famous “classifying tanks” problem would probably not exist if the dataset mentioned in that experiment contained 50 million pictures, not 50 (hoping, that approximately half the pictures were taken on cloudy days and the other half on sunny days). It is this “practice” why all those optical illusions work on almost everyone - because all people’s occipital lobes were trained using very similar data set - of how light works in a 3 dimensional world.
The observer has to be on the higher level
Most people don’t really understand one fact - that our eyes see a 2-dimensional picture and the mind that got trained with all the visual data just perceives and recalculates everything for a 3-dimensional world. If it was not so, your mind would never get confused looking at the infinite stairs.
I also Believe in Ghosts and that you can only ever truly observe something only when that thing is at least one dimension lower than where you are.
Another point worth pondering over is that intelligence in real world is not just about what you already know but more about how fast can you learn new things. The question that came to my mind eventually circled around - if I get repeated stimulus without trying to learn anything (think about doom scrolling), would my brain still start learning? I think, I observe and I say “yes”. There is enough scientific evidence about it, so I am not going to make tall claims here; I don’t have the ground for that. What I do have the ground for is to tell my observations that I have made with myself and people around me. Now, this is not a scientific study, there was no funding for this, and there was no motive either. It is something that has seeped into my mind through everyday life that I live.
The ruin
Human experience is centered around a belief system. One that was fed to our minds early on behaves like the first classifier layer in the neural network. If you were born a vegetarian and stayed one for your first formative years (say, till you became a teenager), it would always be extremely distasteful for you to try eating an animal corpse. A corpse, not food - that’s how the brain will classify it. However, if you somehow started eating meat without being told that what you ate was meat and you liked the taste, would you stop? No. Your stomach, or the rest of the digestive system might say “hey this is not what I am used to digesting” in its own language but unless it causes anything really worrisome, you would continue. Without you knowing, there would probably be no convincing that what you are eating is against your fundamental belief system. After all, umami is in fact one of the basic tastes your body loves to have and meat is where it is in abundance.
The same applies to other parts of our brain. I am past 35 years of age. I come from a time when having a Television was a matter of pride in the neighborhood. I have seen the rise of telephones, then mobiles, then internet. I have made a career out of learning how computers work (on some level at least) and am now witnessing the rise of AI. My brain has accumulated different experiences, which when collected would amount to the “taste I prefer in life”. But all that taste can bloody well go for a toss anyday a crisis hits. I might have to learn new things, new ways of living and dealing with things. I would have to adapt and since I am not a LLM, I would have a fear of death, an urgency, a reason and a strong desire to continue living too. I would try my level best to adapt. If I can’t, I guess death will be able to reach me faster than anticipated. Life might turn bad.
That is enough background for what I want to say. Let me now get back to the point.
Many-a-times I have seen people discussing for and against, with and without proof about whether or not short-form content, especially videos cause cerebral degradation. The term “dopamine addiction” is probably the keyword where you can start researching. It is the term that would summarize all the problems related to over-consumption of short form content, especially videos. But in my own experience, which I cannot back with a study of some sort, that is only one of the many places where the problem lies. Rather I would say that is where the problem began.
Somewhere in 2020, the Indian Government banned TikTok. TikTok probably is the single company with this much controversy around it. TikTok made short-form videos popular, forcing YouTube to come up with Shorts, Instagram with reels and similar stuff by other companies. There were various reasons why governments wanted to ban it. But it definitely made the term “Dopamine Addiction” popular. There is enough scientific evidence that consuming too much short form videos causes an ADHD like behavior in many people. I see that happening with people around me too (including myself). I watch a lot of science-related videos, ranging from space science to computer science and chemistry and more. And even then, I can’t remember most of what I see. I have reduced the consumption of shorts by a lot but I can very well observe that my levels of irritability stays high on a day when I watch more shorts. This holds true for me, my wife, my kid (who is less than 10 years old) and older people around me.
Speaking of “shorts”, I mean “YouTube Shorts”, a feature baked INTO the world’s biggest video platform, uncontrolled and totally allowed by both the governments and the company owning it, without a feature/setting to disable it.
And of course, no one seems to be bothered.
The depth…
Speaking of YouTube Shorts and YouTube - it is controlled by Google. The company whose business model is mostly around “Advertisement”, even after 25 years of existence. I have seen Adsense getting its roots. I have seen YouTube making people rich. And I think it is this “you and I both eat” collaboration between a content creator and Google that makes it so big. I have used Adsense and have earned from it. I like(d) it. But with years in passing, I have seen the slow gradual decline in everything that is out there for consumption.
I am not going ethical on this and am I definitely not asking anyone to follow my steps here but I like reading more than watching videos. If someone wants to read, what does one read? What do I read? I read some blog posts and some news. Some research papers at times, and social media. I think that would cover most of it.
- Research papers are purely technical and pretty rare (once a month or two at max).
- I am away from most Social Media. I closed my insta and facebook account. I don’t post on Twitter, do not engage on Quora anymore and am mostly just on LinkedIn and Reddit these days. Each place is poisonous in its own ways and I try to cope as much as possible with the algorithms trying to get and break (rather crush) my attention for money. I escape and break off when it feels like too much. But that’s my mechanism to cope.
- Blog Posts and News: I have my own ways of getting my tech related content and feed. But most, if not all news sites are the same. Some are different or configured differently but almost all use some kind of off-the-shelf CMS (e.g. Wordpress, Drupal, Ghost etc.). That means they are competing with each other about the same news, for the same advertisement money as any other.
Yes, Advertisements drive revenue, not content. I feel, whether it is a Social Media platform or a website, they are serving any content at all with much contempt and anger. If it were possible, they would just remove all content and keep up the adverts. Google’s policies say that they detect artificial clicks and they do not promote a site owner asking for ad clicks but I think if it were allowed, all the site owners would force you to click a dozen ads without wasting any storage on their servers for real stuff.
We have sites where each page is like - an ad at the header, the content headline (which is often very click-baity), then another ad, then a paragraph about the real topic, then another ad and then another paragraph followed by a “read more” button and then some more ads. A ducking (understand my emotion, there!?) “read more” button for the content I came for!! And the content I did not come for - the ads - they are left, right, center and under the content (those parallax effect ads). What a nightmare!
And since most ad networks place a limit on how many ads a content publisher can serve in a page, each such site has ads from multiple ad networks making things better (oh, only for them, I mean).
…of the ruin
What did I start with? Neural Networks, how we learn and how intelligence is about being able to ingest in new information and adapt fast enough and that something that goes on for longer will leave a longer lasting impact.
When my browser is configured to show (or is not able to stop) advertisements (like when I am using Google app on the phone), I get websites which are totally overloaded with ads. It does not matter which site I am on, if it is related to something recent and is famous enough, I am always going to see ads of all shape (banner, in-post, bottom) and format (animations, parallaxes, picture, text).
Now, like I said - when my browser is configured to show (or is not able to stop) advertisements. Often, when I am traveling or walking, I end up (like a dopamine addict) scrolling news on my phone. With time, I notice that my mind has recognized this pattern that all news sites are going to present me with extreme levels of advertisements and even when my ad blockers are up and there is nearly no advertisements on display, my brain anticipates a lot of distractions and wants to skip down and fast. It is as if my brain has gotten tuned towards distractions and wants to skip to the end fast - even when they are not present. And this is coming from a person who knows that advertisements can be blocked (when most normal/non-tech-savvy people don’t) and has applied that filter on some of his devices.
It was around 2 years that I started noticing this behavior properly. If I stay out of these distractions (don’t check news for long enough and don’t watch YouTube shorts), I do notice that my stability increases, the attention span improves (gets longer) and I am able to learn new things much faster. I have noticed the impact on my elders for the same. I see that people who read or watch lots of news or spend a lot of time on YouTube shorts are generally more irritable. Now, of course people love to be spoon-fed everything and reading takes some efforts. YouTube with its monetization plans has evolved into (or at least is perceived as by a lot of people) a legal get-quick-rich phenomenon. Its content policies and monetization plans, UI, settings, everything does have a special mention of short form content. All that means that there are a lot less people reading anything and much more people actually watching short form content. And those who are actually “reading” are getting all sorts of distractions.
Price and Cost of Intelligence
On one end we have companies, especially bigtech, across the entire spectrum, especially the MMAANG (Meta, Microsoft, Apple, Amazon, Netflix, Google) fame have unleashed advertisements (except Apple, for now; and salutes to them for that) on us from all directions possible and have led the initiatives of monetizing our attention. On the other end, they are using pouring that money to train LLMs which are extremely costly. And from yet another direction, we are being fed food from restaurants (which we do not know in which conditions were prepared) by food delivery apps. These same apps are also doing “quick commerce” and “doorstep delivery of essentials” all while isolating us from the shopkeeper near my house who love chatting with my kid. We are not moving our donkeys and cooking food. Cooking food was once said to be therapeutic. Such things are being kicked out of fashion. All of this because someone who can write clever algorithms decided to cash on where my eyes are and what they see. I am being excited, enticed and invited using clever punchlines, bubbly-bouncy-happy graphics and a barrage of notifications.
I have lately been asking myself a few questions around this entire experience - What are we paying and for what? I guess, the first “what” can we answered with “intelligence, attention-span and brain” and the latter one can be answered with “unhealthy convenience”. I think we need to create a barrier for ourselves to fight against the theft of attention.
Why the hell is my mind being trained for being distracted? With Shorts, with Ad overloads? Google talks about being responsible with AI. What about being responsible with RI, the Real Intelligence?
I don’t expect much from Meta. I expect just a little more from Netflix (and that little exists only because Netflix created an amazing product called “ThePrimeagen”). But I did trust Google, Microsoft and Apple. Apple has been doing ok (thank you, Apple, again 🙏🏻). I don’t use Microsoft’s products much, and even then, I see Microsoft trying to shove AI into me from anywhere possible (like ears, eyes and nose etc.) so I stay away from Windows. But Google!? I thought they were more responsible than this.
I always loved this quote since I read it the first time:
Learning is not free. You have to pay attention.
How, might I ask, am I to learn anything if all my attention has been stolen by systems that want to make me a dopamine addict? I need to fight back. The cost of my intelligence, my ability to learn is the fight that I need to put up. It probably is the same for you too! And no one can do that for anyone else, sadly. So you gotta fight the dopamine addiction yourself; set some rules and follow them like your life depends on it, because from how I see it, it already does.
I can now see the appeal of a Kindle device over an iPad and of a physical book over Kindle. I can now see why I need to walk and talk to the shopkeepers in the nearby stores. I am going to prioritize helping my wife cook food rather than ordering a random stranger knock on my door. I am not sure what all I have lost but I am going to at least try to reclaim which I think was precious and have been losing since last few years.
TLDR
If you jumped here without having the patience to read through it all, you have probably already been ruined. Work harder on reading long texts. Please do. Before your brain turns into a potato. And no, I am not gonna summarize the reasons in this section. Go back where you jumped from and continue reading.
I have to add, in the end, after saying all that I have said, that I, as well as you might be stupid. I just hope that we are not.
PS: I have not run the contents of this article through any LLM. So if it contains some grammatical and spelling mistakes, kindly behave like a human and adjust your brain to process the faults.