Is Object Orientation bad? Is AI driven development bad? Not intrinsically, but anytime people are drawn to convenience, there are hidden tradeoffs they're making.
As far as programming goes, I tend to think that the friction created by bad OO practices haven't really led to anything other than "creative" coping mechanisms in those codebases, so perhaps for this analogy, that doesn't really bode well. But anyway...
Well, LLM's are probably here to stay. I'm not sure that OpenAI has found a viable business model, and as Matt Levine of Bloomberg put it recently, OpenAI is a "money furnace", that takes stupifying amounts of VC money (and, more importantly, cloud compute capacity) and burns it for purposes that do not remotely pay for the cost.
The biggest reason not to rely on ChatGPT specifically, is that it is likely to either disappear or else become much, much more expensive in the future. But, if you can use DeepSeek or some other much, much cheaper alternative just as well, I suppose that will do.
I'm aware of other applications of AI and LLMs, but what most of the people see are the consumer facing ones like ChatGPT/Gemini/etc.
Contemporary novels, especially those depicting modern times, are mostly terrible. I recently read a review of one such modern novel in the Financial Times—-the review was very promising—-and decided to buy and read it. Meanwhile, I am listening to audiobooks or classic, mostly forgotten novels from the last 100 years in my native language. What a difference! One could say that there is a selection effect at work, and that would be fair, but the prose, ideas, and creativity are of such superior quality compared to modern novels that I wonder how and why people read them. Some of the classics are certainly dated, but you can still understand their purpose.
I see the same phenomenon in music and movies, most of which are pseudo-creative works designed to make money in the short term. Movies and music that is quickly forgotten, shared on social media for a couple of weeks and then gone, forever. Although it may be natural to say “kids these days,” I have the impression that the easiest fruits to pick in terms of creativity have been picked in the last 100-150 years, during which more people have participated in creative fields, and in the end, there is not much else to say or experiment with. I mean, one of the most popular film genres today is the biopic, which often features people who are still alive or have recently passed away. In these films, screenwriters and directors sometimes feel the need to tweak certain facts and timelines to make the whole endeavor a little more creative.
I recently commented on a video in which one of today's most popular singers did not sing during their concert, but simply danced (badly, half-naked) with playback doing 90% of the work. Some were surprised by my astonishment, saying that this is how concerts by these new artists are today. That's the vicious circle: people don't even expect singers to sing anymore.
Technology, on the other hand, continues, at least for now, to push the boundaries.
I see it like this: if you weren’t going to be creative in the first place and you’re just grasping for slop to check a box then there’s no loss, perhaps even a slight gain of creativity, if you are fully engaged in being creative you can now prototype things and preview them and spin off ideas that compound and refine and inspire new ones so much faster so the overall creative output is accelerated both “horizontally and vertically” to borrow from compute scaling imagery
Steve Jobs on Crafting Idea to Product
https://www.youtube.com/watch?v=Q3SQYGSFrJY
Tumbling rocks make them shiny metaphor
the_snooze•57m ago
Tech companies want us to be dependent on their information lookup services, while simultaneously not making those services dependable and predictable long-term.