the average developer back then was much better. 50% of engineers in the average IT department produce zero or negative value. I worked at one of the largest financial institutions in the world and had coworkers that literally committed zero code for months.
> coworkers that literally committed zero code for months
That's awesome
I would much prefer to have coworkers that literally commit zero code than the monkeys I sometimes work with who commit way too much code, or just plain commit the wrong code
I don't think it's because of the 'average developer', but because system's like OP's were designed by experienced engineers, while the modern toolchain is written by "senior developers" with 3 years of experience. They have a lot of energy to pull all-nighters reinventing yet another wheel, and not the wisdom to avoid doing that, or to do it in a sensible way.
In fact, the entire tech sector, favouring fresh grads out of college over greybeards, is built this way. My pet very controversial theory is that this phenomenon is massively fueled by everybody and their dog using ADHD stimulant medication to get through their day; fresh grads + amphetamines is the preceding step to Kubernetes and the entire React ecosystem.
tl;dr: modern software is built by 'work hard, not smart' kinda types.
I'm not moving the goalposts. I haven't heard anyone, ever, refer to the Navier-Stokes problem as a top 3 problem in mathematics. People were saying that they thought the solution was in reach a few years ago, before AI was at all capable of research-level mathematics (and the expectation that there was a counterexample).
I am not particularly skeptical of claims about AI, compared to the average here on HN, but that doesn't mean every random piece of hype is warranted. What they did is impressive, even though we now know the only reason they threw so much compute at the problem is that they heard a rumor that someone else was already close. Navier-Stokes is not a top 3 problem in mathematics, and it was the one that was thought closest to being solved.
I have an app with several AI features that I will not mentioning use AI at all because its completely irrelevant and the users will hate it for no reason.
So you recognize that your users will dislike these features. And your solution is to try to hide it, rather than removing the features you know your users won't like? You do you, but that doesn't seem like good product design to me.
im using it for mundane text processing. categorizing and merging recipe ingredients for shopping lists. the users like the feature, unless i tell them its AI. i dont feel any ethical need to disclose that. its just software
So if I was black and racists were using my app that would stop if I put that prominently in the app, i should feel ethically compelled to disclose that?
I think this is because the primary use case of gemini is google search overview and the gemini app. thats probably 98% of gemini tokens. they didnt predict how important agentic coding would become.
Not sure if you'll see this comment since it's been a week, but how did you find this out? Is 3.5 Pro officially cancelled internally? Are they only working on 4 Pro now?
Do labs come back from disasters like GDM’s 3.5 pretrain? I am thinking of Meta’s Llama 4. Meta is just now starting to be taken seriously again but they are definitely not at the frontier. And when I say “come back” I mean have an Opus 4.5 moment, which was really mind blowing for me at the time. Fable was a similar leap, just not as big.
Unless the company is going under, why not? Let's say Google releases Gemini Pro 4 tomorrow, and it's better than Fable and Sol; lots of people would switch over to it.
AI models are almost completely interchangeable, so the best/cheapest/fastest whatever will always have a market.
I agree we’d switch to it. I guess what I’m doubting is if a company can recover from that sort of stumble in the first place.
And they might not want to either. They might think there’s more value somewhere else besides trying to get back to the absolute performance and capability frontier. Smaller models targeted to specific domains that large models would be too inefficient at no matter how large they get or how clever you are at distillation, for example.
reply