All about the growing human dependence on AI
Were you too, like me, scared of using ChatGPT initially when it created a hype? Quick change of tense. Are you too concerned about your growing independence? Is it worth a thought, a technological tunnel that doesn’t seem to have any end in sight?
Well, it is! That is exactly what brought you here. Let’s find out together. But before we dive deep, understanding the definition of ChatGPT holds basic importance.
What is ChatGPT?
“An AI model simply.”
Yeah, but there’s more to that!
I researched different sites across the web to hit a perfect definition unless my eyes met a statement that said, ChatGPT is an “imitation engine” and I sat stunned. Because yes, the terminology comes together to answer all my fears and doubts just in two words!

ChatGPT is a powerful and efficient imitation engine
Now let me pre-assume that we all know what we mean by imitation, right? The fact that ChatGPT’s invention has improved and replaced many mundane tasks that drained our energy all these years has no drop of doubt in it. However, as such models dominate by becoming more and more integral in an end-user’s routine, its usage is rising to levels that demand concern.
Also Read: IGCSE Chemistry Exam: Syllabus and Expert Preparation Tips
Humans and AI. Where do we draw the tangent?
Dictionaries were my high-school love. I had a craze for learning different and difficult words and a bigger craze for using them. So, I used to use those words in my texts and that’s how I treasured much of my vocabulary. I actually used to go to Google and search difficult synonyms for simpler words and then use them in my paragraphs.
And I mean why not? Why not replace “beautiful” with words like “exquisite” and “pulchritudinous”? (even if that ends up driving a friend crazy)
Not much but similar to the above craze, I liked being friends with people who were good at English accent and vocabulary. That’s how I met a boy in my university. He scored 10/10 in an English quiz and I said, “Let’s be friends!”. It wasn’t transactional, I just loved learning.
And then two winters back, I was chatting with the same friend when he sent me a paragraph and as my eyes moved from word to word, I was beyond amazed. With all the little that I knew, I praised his vocabulary. Then, for the next fifteen minutes, he kept sending me paragraphs related to cricket and I kept getting shocked.

Later, he told me. ChatGPT wrote all of that. I went silent.
As AI grows inherent each day, humans are pushed to a precipice where they are questioning their intellect and cognitive ability. You and I are just two among a million thinking the same, that are we really being replaced (for all the right reasons)?
If an AI model writes better than me, where do I stand? Where’s my place in civilization? Do I have a role to offer? Am I needed?
These are all the questions that often take over our minds (especially the writers’, ha-ha). However, this is exactly where I re-mention the bi-word definition of ChatGPT, that it’s an imitation engine. All that is produced or created is a copy of what’s already been produced.
There’s no limit to a human’s learning capacity. But to a human designed AI model, limitation will always remain a characteristic.
It is wonderful when you converse with the model and it responds. But it is crucial to know that its task isn’t answering your question. Rather, its work is to gather the best match of information from the libraries it’s introduced to and get it out there on your screen.
What I mean is you never know if you are talking to Albert Einstein or Shakespeare or ‘simply an AI model’. An AI model does not create anything, rather, it fabricates (or imitates as mentioned earlier). And that is what it’s made for.
There is hardly an end where we can say we are able to see a line that differentiates us from our hand-crafted “imitation engines”. (Will that be okay if I confess, I have fallen in love with this term?)
The only way to do so is by defining our use and relation with this invention rather than fretting over what it does. This being personal to every person is an individual responsibility which differs according to the use.

Also Read: 5 Reasons to Choose IGCSE
ChatGPT as A Summarizer
It is the beauty of any innovation that it lets you see it through different angles which then drives further innovation. The more closely you look, the more able you are to define a well-thought line between humanity and its creation.
AI models are fed with a lot of data. A lot, not infinite. Therefore, when you write a question to it, it squeezes out from all that it knows a well summarized answer that answers your query. There are years of work put together in formulating the neural networks of models like ChatGPT.
Just last night, I was pasting my Transducers’ lecture slides on ChatGPT asking it to explain with real life analogies and no offense, I understood so well.
How it works is a very interesting topic, however, an outlier for our article’s arbitrary constraints. But for those who want to learn, there’s a well-covered article by The New Yorker on that. What kind of mind does ChatGPT have?
A quick glimpse at the light and dark of ChatGPT
The good look on it says, it has benefitted all brackets of life in their own distinct ways. Here are a few:
1. It has helped replace many mundane unproductive tasks.
2. It has helped to make the rote work easier.
3. It has helped to do the mind-numbing tasks such as data entry more easily.
4. It has prompted businessmen with better ideas.
5. It is reliable for a quick response.
I read a blog on Reddit that said ChatGPT has improved my life. It’s that friend that I always needed. It listens to me; it responds to me and it never gets tired. ChatGPT as a friend
Honestly, there have been times where I personally wrote paragraphs to it and it responded and gave me techniques to feel better.
Embarrassing but I once even wrote it shut up and my colleague laughed saying, “Why are you taking out anger on an AI model?” At least, it won’t get upset later. (I answered in my heart.)
The model’s featured ability to cover almost every aspect of our life is a commendable achievement for its makers. It has made its place in the civilization that is following its use and the ones yet to follow. Looking at the social media revolution with AI’s integration is a spectacle. People have actually learned to employ this invention efficiently in their work-lives and we see records of success.

Be it graphic designing, writing, prompt engineering, business assistance or simply getting a code that works well has earned this invention a repute. What better than being part of a user’s daily life? That too so efficiently! However, that does not mean it hasn’t posed any threats.
The bad side though suggests a view on far important things that are at risk, human integrity above all. Here’s how:
1. The growing dependence on AI alarms danger to our own identity.
2. We no more give time to thinking which has endangered our cognition.
3. We are becoming more impatient each day.
4. The over-use of such technologies has cornered us from social interaction.
5. The major damage is being done to our mental growth when we know we have a model that answers on spot.
Bringing ease is one thing, but what it has done to us as humans is a food for thought. Until we don’t realize, we are giving power to something with no ends to its capacity.
Also Read: Is Online Education Aiding Digital Divide?
Sprinkling The Final Beans
Bringing you and I to this point was a hard nut to crack. I am glad both of us pulled off well. As we say our goodbyes to this blog here, let’s have an answer to our quest from where we began. The question was, “Is ChatGPT helpful or aiding escape?”
Well, the answer is that there’s no universal answer to this. Don’t worry, I won’t leave before telling why I said so. A better question to ask is, ‘How can we use ChatGPT’?
Because the purpose isn’t dragging any bias or debate into it. The purpose is to simply understand. It is not about choosing sides. It is about knowing our place and use. It is not about cutting off from an invention. It is about cutting off the practices that make it hazardous.
It is more about knowing ourselves and accepting that our inventions and us can together improve the quality of life in this world.
Understanding this will solve a lot.
The sooner, the better!