Posted by mips_avatar 15 hours ago
AI the manual algorithm to generate code and analyze images is quite an interesting underlying tech.
Copilot for Word. Copilot for PowerPoint. Copilot for email. Copilot for code. Worse than the tools they replaced. Worse than competitors' tools. Sometimes worse than doing the work manually.
This is revolting. Three years ago I’d have said this is a terrible black mirror plot
I hate the entire premise even though some of it has been useful but at worst you're creating code and/or information that's just wrong and "can get someone killed" (metaphorical, but also probably literal), you're creating absolutely unrealistic expectations
Zuck said they'd be able to replace engineers with AI. Well, that tells you everything you need to know, doesn't it? With all of the scandals Facebook properties have had over the years. A real engineer/competent CEO wouldn't say that
Not only because it's destroying creator jobs while also ripping off creators, but it's also producing shit that's offensively bad to professionals.
One thing that people in tech circles might not be aware of is that people outside of tech circles aren't thinking that tech workers are smart. They haven't thought that for a long time. They are generally thinking that tech workers are dimwit exploiter techbros, screwing over everyone. This started before "AI", but now "AI" (and tech billionaires backing certain political elements) has poured gasoline on the fire. Good luck getting dates with people from outside our field of employment. (You could try making your dating profile all about enjoying hiking and dabbling with your acoustic guitar, but they'll quickly know you're the enemy, as soon as you drive up in a Tesla, or as soon you say "actually..." before launching into a libertarian economics spiel over coffee.)
I wouldn't shit talk you to your face if you're making an AI thing. However I also understand the frustration and the exhaustion with it, and to be blunt, if a product advertises AI in it, I immediately do treat it more skeptically. If the features are opt-in, fine. If however it seems like the sort of thing that's going to start spamming me with Clippy-style "let our AI do your work for you!" popups whilst I'm trying to learn your fucking software, I will get aggravated extremely fast.
Except it didn't stick? https://news.ycombinator.com/item?id=43088369
Well, it's not the fault on a random person doing some project that may even be cool.
I'll certainly adjust my priors and start treating the person as probably an idiot. But if given evidence they are not, I'm interested on what they are doing.
I'm all for shaming people who just link to ChatGPT and call their whatever thing AI powered. If you're actually doing some work though and doing something interesting, I'll hear you out.
1) A third party app simply cannot compete with Google Maps on coverage, accuracy and being up to date. Yes, there are APIs you can use to access this, but they're expensive and limited, which leads us to the second problem:
2) You can't make money off them. Nobody will pay to use your app (because there's so much free competition), and the monetization opportunities are very limited. It's too late in the flow to sell flights, you can't compete with Booking etc for hotel search, and big ticket attractions don't pay commissions for referrals. That leaves you with referrals for tours, but people who pay for tours are not the ones trying to DIY their trip planning in the first place.
So many products are like this - it sounds good on paper to consolidate a bunch of tasks in one place but it's not without costs and the benefit is just not very high.
If they become popular they'll have to move to OSM, Google's steep charging for their Maps API at high usage that has brought companies to their knees is well known [1].
Similar to "made for everyone" social networks and video upload platforms.
But there are niches that are trip planning + there are no one solving the pain! For example Geocaching. I always dreamed about an easy way to plan Geocaching routes to travel and find interesting caches on the way. Currently you gotta filter them out and then eyeball the map what seems to be nearby, despite there, maybe, not being any real roads there, or the cache is probably maybe actually lost or has to be accessed at specific time of day.
So... No one wants apps that are already solved + boring.