5 ms·
Having engaged in a couple previous hype cycles (2017 crypto, 2020 tools for thought, 2022 Twitter alternatives), this one feels qualitatively more mainstream,
by ssalka 3y ago
Having engaged in a couple previous hype cycles (2017 crypto, 2020 tools for thought, 2022 Twitter alternatives), this one feels qualitatively more mainstream, and thus overall larger - it's not just tech bros getting into these tools, it's virtually everyone. So I tend to agree that we're going to see a big AI boom in the coming years.
My main question at this point is whether regulation and/or lawsuits will hamper any of that progress or at least slow it down.
- yCombLinks 3y agoAI feels far more like the internet hype cycle of the mid to late 90's and the smart phone hype cycle of the mid to late 2000's than anything you listed. Your examples didn't make much noise outside of niche interested audiences. AI hype is everywhere.
- jvanderbot 3y agoThis one feels more than all those but less than web 1.0. it'll be big for a bit, but it can't do everything everyone wants.
- ryandrake 3y ago> this one feels qualitatively more mainstream, and thus overall larger - it's not just tech bros getting into these tools, it's virtually everyone. This may just mean the bubble is inflating faster than with the previous examples. Mainstream excitement often = it's about to pop. The time to sell high and bail on crypto was when it became mainstreamed in the general public. I think the exact day was when Elon Musk hosted SNL and mentioned Dogecoin. The time to bail on Real Estate was when the receptionist in your office quit to become a Realtor. The time to bail on dotCom 1.0 was when your dad had a web site idea.
- zamnos 3y agoDogecoin is down from the 2021 high but this Internet thing seems to have had legs. AMZN is up at $113 from shy of $4 circa 2001. Sounds like your dad's one for one. What's his thoughts on ChatGPT?
- oxfordmale 3y agoIn a hype cycle people overestimate the short term impact and underestimate the long term impact. Developers won't be redundant, however, sites like Stack Overflow may struggle. There will be job losses, but not in the way people expect, however, there will also be job gains in other areas that were previously uneconomical.
- andrei_says_ 3y agoWhat are some areas dependent on LLMs where you see job gains proportional to the losses related to LLMs?
- oxfordmale 3y agoThe good equivalent example is translation. You would expect Google Translate to make translators redundant. However, they are busier than ever. As they can now translate documents a lot faster, it becomes more economical for companies to translate documents. Regarding job gains, we initially will see a whole bunch of snake oil sales people selling AI.
- pcthrowaway 3y agoErr.. but AI has pretty much obsoleted most translation work. Of course, well-written poetry and prose will still need to be translated by an expert in both languages, as well as possibly literature and linguistics. But even then, GPT-4 may be better than this person some of the time.
- oxfordmale 3y agoAbsolutely not. OpenAI can't be taken to court if a mistranslation causes issues. That is why it will always require a human to confirm the translation is correct. As discussed before, OpenAI doesn't know it doesn't know, and will dream up something.
- quickthrower2 3y agoGonna be prompt engineers (and human langchains)
- quickthrower2 3y agoI have heard completely non computer people mention it. Whereas the other things never.