i feel like we’ve been ‘a few months away’ from ai replacing all devs and designers for like a year now

Check developer threads on Reddit, Levels or HN, for example. Even many senior engineers with years of experience have found the market to be tough.

Didn't say it was the only factor but it's a really big factor. You wouldn't fire people you need. Even if they're expensive. The truth is that even things like Copilot have 10x developer productivity. This is according to my developer friends themselves.

My point was that o3 is a whole new level of AI. With Copilot (which has already 10x'ed dev productivity) you still need to really know what you're doing. o3 will likely allow anyone who has a product idea create a product. No years of tech expertise necessary.

i would agree it makes just starting to create a product more accessible but this is different than it’s ability to outright replace workers for a big tech company, for example.

workers as in devs, in the above post^ this is from the perspective i have developed working on ai and automation tools in an engineering/tech lead role day to day, fwiw. these are tools that can disrupt no doubt, but the impact they currently have is often overstated imo.

Currently, yes. As I said, Copilot helps an engineer be much more productive but you still need to have engineering skills and experience. o3 sounds like a completely different beast. We'll know more soon.

my experience and perspective wrt tech industry is that there was genuinely a lot of excess for a while (when i entered uni the idea of graduating into a cushy tech job was almost a given) and as times have gotten tougher this excess has dried up.

there’s still a lot of tech jobs i see, but the bar to entry is higher and the competition is much fiercer. compared to the state of hiring in tech industries not even a decade ago i think this amplifies people’s perspective of how poor things are.

++ i also agree with what you’re saying, Jamie

The free money well has dried up and people are scrambling for new ideas.. So they're all jumping on the AI ship whether or not it makes sense to do so 🤷🏻‍♀️

Crazy how much credit people will give to text predictor++, even though it still doesn't have an understanding of what the problem you're asking it is. It's helpful, sure, but I don't think most people are losing their jobs to this just yet

Especially not people whose entire job is to think and reason about problems logically. If all you do at your day job is press buttons on a keyboard like Employee 427, then yeah LLMs can probably replace you, because you're not doing much more than they are

But if you have to think about a difficult problem, and you have to reason with the given conditions and constraints, you won't get very far with just prompting, because oftentimes your problem will be more complex than whatever it's been trained on

I also don't know how you can trust research from the company that's trying to sell you the product. Of course they're going to make it look good, they want to make money. That's how marketing works. It's like putting all your trust in the "9 out of 10 doctors" thing

That's the great thing about this tech. People will be able to test it independently and confirm or question what Open AI said about it.