AI reminds me of early COVID in that it's difficult at times to tell how much it will completely change everything and therefore how concerned to be.

And also in the way that it's hard to tell what we as the average people should do to prepare and about it.

And how!

Like any tech business/product LLMs rely on constant presence. Posting low value "Look Ma What I Told the LLM to Do!" articles keeps it in the spotlight. Ignorance is bliss however.

It feels like all the crypto hype but to an even more extreme degree. At least with all the blockchain bullshit you could be confident that all of the breathless announcements about decentralized ledgers for whatever where going to be quietly abandoned when it turned out they didn't do anything except get dumb investor money.

The main thing I mind is how products (primarily G-suite with respect to which I realize I'm actually the product ) are riddled with ads for it. My productivity gets destroyed by distractions and now every time I pause to think or load a new page, I get a "get help writing" popup or an ad for some other useless AI feature nobody wants.

It's like Poochy on the simpsons (might be too old a reference) - everyone hated him and Homer suggests that when Poochy's not around, everyone should be asking "where's Poochy?" I hope these useless features suffer the same fate (Poochy dies).

I can filter out the dumb "Llama 3.4 changes everything" posts, but the dumb ads, for ostensible productivity features (that everyone knows aren't) are destroying my productivity.

There is reason to be worried and excited. I would say web3 and crypto were hype. The VCs bet on it heavily and tried to make it happen, but it didn't.

AI is still in its early stages and is already disrupting jobs and work. Many enterprises that were laggards in technology adoption are now embracing AI use cases. Every week breaks an assumption that was made in the previous one. It's all tiring and demanding to keep up. Perhaps there is some hype surrounding agents and the extent to which they can deliver, but the big bet is that upcoming models will outperform their predecessors.

From a b2b perspective, agents are already making a difference in enterprises. So there's substance behind the hype.

I think most people are, although there's some interesting stuff beneath the hype.

Yup completely sick of all this and all the harm it already produces, be it environmental or webhosting where they DOS you server scraping everything on it, or fake "AI" generated bug reports and other spam for FOSS projects or all the money it sucks out of really useful and really socially beneficial projects where money (and other resources, including attention) would be much smarter spent on. I hope all this "AI" or annoying idiocy hype crashes soon.

I think a lot of the hype has to do with the fact that since English has now become a programming language, everybody can theorize about what might be possible without any other programming or mathematical knowledge. This naturally results in a lot of noise, as random ideas by laymen and genuinely great, currently possible ideas now have the exact same form and even language. To outsiders, a good prompt might appear like it is good feasible idea. Only with background knowledge can you judge if an idea is currently feasible or if it will likely become feasible soon.

Still I think instead of getting tired of it being in the public spotlight now, it makes sense to just upgrade your inner filters just like we unconsciously ignore other noise as it is much more than hype. As Karpathy said, it will eat through both traditional software and neural networks. Imagine when he introduced the term software 2.0 [1] and someone kept doing everything in C++ as they didn't want to "buy into the hype".

On a side note: I really like Karpathy’s evolving software concept, especially since these categories weren’t just invented now to explain genAI's place. Rather, genAI has now been assigned as software 3.0, when his distinction between software 2.0 and software 1.0 was done already in 2017

1: https://karpathy.medium.com/software-2-0-a64152b37c35

There is a very clear pessimist case for the current crop of Large Language Models where:

1. training costs are going to continue to escalate much faster then the models improve.

2. the current paradigm of trained ahead of time token completion turns out not to be a pathway to "AGI" or even anything much more then we have now. As this becomes clear investment dries up.

3. without investor support user's are forced to pay full price for LLM queries and a lot of things that are super useful when this shit is subsidized up the wazoo are not worth it at full price and when people are given the choice of stopping using LLMs or ponying up more cash people realize they can get by without them.

4. Everyone who was hyping LLMs moves on to the next shiny thing and all money for "AI", even for useful stuff like old fashioned machine learning, dries up and we end up somewhat backwards from where we started except GPUs are cheap again assuming the crash doesn't take down NVIDIA.