5 ms·
The issue really is that the AI isn’t good enough that people actually want it and are willing to pay for it. It’s like IPV6, if it really was a huge benefit t
by daft_pink 1y ago
The issue really is that the AI isn’t good enough that people actually want it and are willing to pay for it.
It’s like IPV6, if it really was a huge benefit to the end user, we’d have adopted it already.
- immibis 1y agoEnd users don't choose ipv6 or not - ISPs do
- NitpickLawyer 1y ago> isn’t good enough that people actually want it and are willing to pay for it. Just from current ARR announcements: 3b+ anthropic, 10b+ oai, whatever google makes, whatever ms makes, yeah people are already paying for it.
- meheleventyone 1y agoGiven everyone and their mother is putting AI in to their products it makes me wonder how that revenue breaks down between people incidentally paying for it versus deliberately paying for it versus being subsidized by VC. Obviously ultimately all this revenue is being collected at a massive loss but I wonder if that carries on down the value chain.
- squidbeak 1y agoAmusing the way the argument shifts every time. This one's new though. "If it was any good, people would pay for it." "The data shows people are paying for it." "Aah but they don't know they're paying for it."
- watwut 1y agoThey have been multiple crashed again and again due to people bot actually paying. And VC investments are distorting markets - unprofitable companies kill profitable ones before crashing.
- meheleventyone 1y agoI don’t think I’m trying to make that argument but thanks for putting it in my mouth. I do pay (or via employment get paid access) for a lot of products that have AI features that I don’t care about so from personal experience I know that at least some of the value chain is incidental.
- supersparrow 1y agoHuh? I’ve been programming for 20 years now and LLMs/GenAI have replaced search and StackOverflow for me - I’d say that means they are pretty good! They are not perfect, not even close, but they are excellent when used as an assistant and when you know the result you’re expecting and can spot its obvious errors.
- ethan_smith 1y agoIPv6 adoption is actually limited by network effect and infrastructure transition costs, not lack of end-user benefits - unlike AI, which faces a value perception problem.
- brookst 1y agoChatGPT has more than 500m DAU, three years after creation. Is that really a value perception problem?
- nonplus 1y agoThat value (of one company) is from speculative investment. I don't think it negates that the field has a perception problem. After seeing something like blockchain run completely afoul/used for the wrong things and embraced by the public for it, I at least agree that AI has a value perception problem.
- brookst 1y agoHow does speculative investment get 500m DAU?
- bitwize 1y agoI dunno, how did Juicero get $120m?
- ryandrake 1y agoExactly. It is possible for a metric like DAUs to come almost entirely from marketing saturation, heavy promotion, and hype, and not from actual utility to the user. I'm not sure that's the case in particular for ChatGPT but I wouldn't be surprised.