The Generative AI Con

But that still doesn't exclude copilot, which is the most intuitive and genuinely useful version of ai after chatgpt itself.

Hey I have my gripes with the landscape, certainly, but this is just too much.

> It sure is! But it doesn't really prove anything other than that people are using the single-most-talked about product in the world. By comparison, billions of people use Facebook and Google. I don't care about this number!

> User numbers alone tell you nothing about the sustainability or profitability of a business, or how those people use the product. It doesn’t delineate between daily users, and those who occasionally (and shallowly) flirt with an app or a website. It doesn’t say how essential a product is for that person.

Both of these "arguments" could be applied to any of the big tech giants of the last 25 years - Google, Amazon, Facebook, Uber, whatever (and there'd be other incumbents used by billions of people before them). I don't believe these arguments discount ChatGPT from having the potential to continue growing like a Facebook. And who cares how many journalists Altman knows, you don't get a product written about that much unless it's truly a groundbreaking product.

> And even then, we still don't have a killer app! There is no product that everybody loves, and there is no iPhone moment!

There sure is, it's called programming. He called out quality earlier on, but the quantity and speed and direction the AI can take (as well as its rate of improvement) is breathtaking. My own output has 10x'd easily since GPT-4 came out (although some of that means I'm needing far less hours in certain places). And guess what? The code quality is generally fine.

> Where are the products? No, really, where are they? What's the product you use every day, or week, that uses generative AI, that truly changes your life? If generative AI disappeared tomorrow — assuming you are not somebody who actively builds using it — would your life materially change?

Ok, the product is called ChatGPT, or Claude, or DeepSeek or whatever, and if it disappeared overnight, my programming productivity would drop dramatically. I would not seek to take on as ambitious projects in as short of a time frame as I am doing now.

I don't know, as a user and developer both of AI/LLMs, this article isn't hitting the mark for me. There are legitimate criticisms of the field, but I'm not seeing them thus far.

Edit - I'll say I agree with the Deep Research criticisms. These products are very underwhelming. They're literally to help people do a research report which needs to be done, but won't be used or read critically by anyone report.

By this I mean it’s a bet on what R&D might yield, current progress being some kind of signal. No one has certainty here. It’s an emergent technology and no one knows for certain how far it can be pushed.

LLMs are pretty good at the aspects of coding that I consider to be "the fun part". Using them has made me more productive, but also made my job less fun, because I can't justify spending time using my own brain to do "the fun part" on my employer's dime. And that was something I was particularly good at, which is why I was able to be paid well to do it.

So now my company makes more money, and the work gets done faster, but I can't say I feel appreciative. I'm sure it's great for founders though, for whom doing the work is merely an obstacle to having the finished product. For me, the work is the end goal, because I'm not hired to own the result.

He might be. It’s an emergent technology and no one knows for certain how far it can be pushed.