A Philosophically Technical View Into Generative AI

Conclusion

When something has had fundamental effects on a field, you have to start questioning the fundamentals of that field. The overhaul in the industry (especially the tech industry) is not something to take lightly, as you might miss a lot of things because you're among the first people experimenting with this completely new workflow. AI is a very useful tool, but useful doesn't mean safe; a knife is useful and unsafe at the same time.

The concept of "seniority" has always been connected to experience. But now we gotta question the fundamentals: How are experiences connected to knowledge? What makes experiences experiences in the first place? What's the difference between an experience and a memory?

One of the things that seniority has taught us as software engineers is that "if it works, don't touch it." It might sound stupid, but it's just a funny way to explain a very important concept: changes are risky. Every time you change something in your system, you're introducing new ways to break things. The larger the surface area, the riskier the process becomes. But now everyone wants to do software development in a completely new way, using indeterministic tools that hallucinate and that we've only been using for a few years or even months. This is the harshest change ever, but some people are surprised that you worry about it, as if the seniority you gained had been a hallucination all this time.

AI is definitely capable. It has helped us do things beyond our wildest imaginations, and I am by no means against it. But using AI as a new tool is one thing, and using AI as a way of gambling with your future is another. This... thing that AI is doing is nothing like the tools I used to know. It's more like a constantly drunk employee who also forgets everything every few hours.

Software engineering is among the jobs that have changed the most. We, as programmers, are used to changes in our industry much more than most other professions. We are completely okay with replacing our "tool" with "newer tools." Yet the conversation around AI in programming is very loud and dramatic. Even people with the most flexibility aren't okay with this sudden rush of changes, and quite literally everyone—including those who have wholeheartedly accepted AI in their jobs—is saying that they're having substantially less fun doing what they did one year ago. The constant speed of everything is absolutely overwhelming.

Maybe I'm wrong. Maybe this whole book is just the bargaining stage of my 5 stages of grief. But it's not precisely the best time to be a software engineer. Even if you figure AI out, as I feel like I've just done, you still haven't changed the reality around you or the job market's expectations of you. The contractor doesn't care what tool you use, but they ask you to deliver the product in such a short amount of time that it's practically impossible to do it without offloading everything to AI.

Your job was already full of stress, and now they want to force you to get rid of all its meaning and joy. If you want to keep your job, you have to conform to their new policies, which set you up for failure without you even knowing it. The policies force you to forget who you are, bury you under prompts and tickets, keep you from feeling productive, and make you miss the flow state you used to be in. The funny thing is that staying in this situation would definitely degrade your expertise, giving you even more FOMO. But if you mention this, they just advise you to "go learn," and they never tell you how or what to learn.

We just wanted to become good programmers—to get better at our jobs, become mid-level developers, senior developers, better seniors, principal engineers, and so on. But now everything has lost its meaning. You're constantly producing with no sense of progression. We don't know what to learn. Putting yourself in the position of learning the fundamentals seems so damn hard and pointless at this point, when you know AI can do in a few seconds what you're merely trying to learn.

Who knows what will happen? This is the wildest decade we're experiencing—starting with the COVID pandemic and moving straight into the age of artificial intelligence. And for what it's worth, this is the beginning of a new era: the era of burnout.