Artificial Intelligence: Double Edged Progress
- Jishant Acharya
- Mar 1
- 4 min read
If you’re not living under a rock, you’ve probably heard about artificial intelligence. The new kid on the block that’s impacting everyone’s jobs and has caused companies worth more than the GDP of some countries to see their market caps fall off a cliff. We’re in an active revolution led by data and it’s mind-boggling how much we can do with it. It’s not just about how powerful the technology is; much of the credit also goes to the people creating practical use cases that make adoption easier. The advancements aren’t just focused on the tech industry; they’re leading because developers want to reduce the work they’re supposed to do before thinking about anything else. The advancements have now reached most sectors; they have reached a point where many jobs have become obsolete. If an AI can help you manage support requests, talk to people, and do many of the things humans can do, and it works 80% of the time, that’s not insignificant. We had human errors before; now we’ll have intelligence errors cropping up everywhere.
If it’s so good, why is it seen as a double-edged sword? The reason lies in human resources and the ability to tame the beast - something only a few will possess in the near future. I’ll explain it from development, financial, and social lenses.
Development
Take the example of a person who is vibe-coding a huge and complex application that they have no real understanding of. Let’s say they got lucky and, in the 80% of cases where AI works, it worked for them. This person built an application, sold it to a bunch of people, and is now trying to scale it successfully. When the application then picks up traction, the person tries to add more features with the same confidence. This time, the AI falters and creates an issue that’s very difficult to catch, and it ends up in production code. Let’s say it fails in a way that exposes the application to cyberattacks. The person doesn’t fully understand what’s happening and ends up becoming part of an incident. This shows how important building things is, but also highlights the added importance of understanding what we’re doing. This is the development side.
Economic
Now let’s deep dive into the economic side of things; the above illustration is a very simple example of what’s wrong with the current revolution. As you can see, the money is moving in circles, and there is no real money being generated. AI needs huge investments, starting with training, then maintenance, and finally deployment at scale. It requires large amounts of RAM, compute resources, and GPUs to run. There has been a lot of talk about how much OpenAI is earning and could earn in the near future, but if that’s the case, why is it introducing ads on its platform? The reality is that the way things are progressing may not be sustainable, as the cash burn required to maintain such a giant is extremely high.
So, where will the real money be generated from? It will come from real AI adoption and people paying to use it. Initially, adoption was driven by fear of missing out. A lot of companies rushed to push AI adoption, encouraging employees to use it aggressively. Many positions became redundant, leading to layoffs of mammoth proportions. If the current level of burn is high because of maintenance, and the very people expected to fund it are out of jobs because of it, where will the real money come from? On top of that, the narrative that AI can replace the work of five to ten people, depending on the level of specialization required, is starting to fade [1].
There is also a wave of rejection as many companies haven’t seen any impact. To put salt over the wound, there are hilarious stories where many production databases have been cleared out by the AI in the name of doing things where the person using it tried to show full trust in AI. Many say that the code generated is not maintainable. Financial analysts cannot depend on the numbers it gives. Additionally, support roles are being overwhelmed because AI wasn’t able to support their customers as needed, and escalations are being raised constantly. Many of the jobs that were said to become obsolete are now popping up on company career sites to hire humans back. In the end, there is also an anger factor towards the very thing that has taken their jobs away. Why would any sane person fund themselves out of a job, the very thing pivotal to their success in life.
Social
Every revolution comes with social factors too - the White Revolution, the Green Revolution, and the historically significant French Revolution. So does this one. This one is about the loss of thinking; I’m a developer, so I’m naturally biased toward examples from my own field. Gen Z is the first generation reported to have lower IQ scores compared to the generation before them, which is a significant development. Attention spans have also been reduced by a constant stream of short-form vertical content on apps like TikTok, Instagram, and others. The attack on thinking capacity is happening due to AI, as the newer generation became early adopters of technology geared toward making their lives easier after school and after all the basics are cleared. It is human nature to take the path of least resistance. It is human nature to take the path of least resistance. Now, what’s happening is that they can build a multimillion-dollar SaaS platform but don’t fully understand how it works.
Conclusion
AI is a boon to productivity. The amount of work it helps accomplish is mind-blowing, and its accuracy will increase over time as models become more distilled and continue learning from us. To stay ahead, remain employable, and not become just a collection of bones and flesh, we need to find ways to ration the use of AI for upcoming generations and counter its social effects. That could be monumental in creating a domino effect that helps us control and tame the use and growth of AI.
Citations:
