Working On Borrowed Time
I felt like I was fighting a loosing war, I couldn’t speak about it almost had me on a chock hold, where there where no words to be said. I’ve allways been mesmerized about the Idea of AI, AGI and topics under data science. It shifted to my once mechanincal engineering mindset to shift to a more data accelerated mideset, so I joined my computer science studies around 2023 in the hopes of one day joing Data Science. It was a year where companies were chasing this new gold rush, Large Language Models(LLMs). The realase of ChatGPT seemed like a marvel of research and also taboo to the old school development communinty. It had caused a battel between these two factions. I was unfortunate to get caught in the crossfire. Ohh boy it’s just the start of it
The Golden Days 2023
It was not all doom and gloom it had brought many new job opportunities, data analysts, data scientists, data engineer, worforce on data centers construction, setup and data labling. It was not only the big coorporation that rushed for the gold but it was also countless individuals in this feild.
Current status
Research shows that from the first models to the recent models Hallucination rates have decreased by 3% every year, making the ones neanderthal of LLMs to becoming these powerfull tools and workforces. Giving this insecurity that any task realted to text was going to be replaced with AI. That was not the only matter OpenAI and Anthropic released what is said to be the “Entry Level Jobs” for people in tech. The relization hit that something is creeping it’s way towards, actually it was here already.
What Made me anxious
Is a qoute from Matt Shumer CEO of HyperWriteAI and OthersideAI, on his recent viral blog said.
I was using AI more and more, going back and forth with it less and less, watching it handle things I used to think required my expertise. Then, on February 5th, two major AI labs released new models on the same day: GPT-5.3 Codex from OpenAI, and Opus 4.6 from Anthropic (the makers of Claude, one of the main competitors to ChatGPT). And something clicked. Not like a light switch… more like the moment you realize the water has been rising around you and is now at your chest.
That when I just couldn’t comprehend someone that level speaking with such certaintiy and authorithy, that was the choke hold. Lost count of time, heart rate rose and I couldn’t breath. I took me back months earlier, to a conversation that I was having with my Ferese driver(a private transportation service like uber) a very wellspoken person. It all started around the time where AI music production had been headlines of news outlines. He said along the lines of AI is outright wrong, I totaly understand from what point of view he was talking. He had grown on a time where art was soly connected to the artist. He even brought up auto tune as a technology new to their time but despised by artist who honed their vocal for years to be bested by someone technical. I retored saying that AI is for the greater good, brushing away the alarm that he sounded.
What the future of us text based feilds
What the progress realy about
Open source or hidden behind a supscription fee, these agentic AI really accelerate workflows, which myself seen through my experience. They’re making people thousands of dollars and there giving a fighting chance to start up. Even if these powerfull systems are out their I don’t think there are thier yet, I think entry level jobs are still safe. But for the foreseeable future they won’t sit and wait for us.
What can we do
I know most people’s problem stems from the exsustion of keeping up with every new release, every new topics in data science. That’s what challenged me first, that was all the same when big javascript frame works ” which one should I learn ”, ” I need to know everything “. Causes burnout and loss of creativity, it all about focus. Choose a niche feild then expand from there. I think that is what really helped me.
Conclusion
AI can write novels, songs, program and anything text related but the art is soley connected to the artist. On a last note Matt Garman, the CEO of Amazon Web Services (AWS). In late 2025, he strongly pushed back against the idea that Large Language Models (LLMs) would replace junior developers, calling the strategy “one of the dumbest things I’ve heard”.
His reasoning aligned with the idea that AI cannot truly understand intent, context, and the nuance of what a human “wants to say” in terms of business requirements.