Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The trust collapse: Infinite AI content is awful

arnon.dk

Post reply on HN
Print view View original

71–80 of 218 posts

Re: The trust collapse: Infinite AI content is awful

#71

Post by SeanDav »

everdrive wrote: post #7

Doesn't matter. We must keep building more and more technology no matter the cost. Have an idea for a business? Build it. Does your business make the lives of people worse? Doesn't matter, keep pushing. Could some new technology ruin the lives and relationships that people have? Doesn't matter, just build it. We always need more, need to do more. Every experiment is valid, every impulse must be followed. More complex…

One of the potential upsides to this as that people just might start taking time to engage in a bit of critical thinking before reacting. Is this real? How likely is this AI nonsense? What is the source? Is this the full picture? etc.

Perhaps I am too optimistic...

Re: The trust collapse: Infinite AI content is awful

#72

Post by kyoob »

heddycrow wrote: post #51

I wish we were talking about what's next versus what's increasingly here. How can infinite AI content be strictly awful if it forces us to fix issues with our trust and reward systems? Short term, sure. But infinite (also) implies long term. I wish I had a really smart game theorist friend who could help me project forward into time if for nothing other than just fun. Don't get me wrong, I'm not trying to reduce the…

From a game-theory perspective, if players rush the field with AI-generated content because it's where all the advantages are this year, then there's going to be room on the margins for trust-signaling players to advance themselves with more obviously handspun stuff. Basically, a firm handshake and an office right down the street. Lunches and golf.

The real question to ask in this gold rush might be what kind of shovels we can sell to this corner of hand shakers and lunchers. A human-verifiable reputation market? Like Yelp but for "these are real people and I was able to talk to an actual human." Or diners and golf carts, if you're not into abstractions.

Re: The trust collapse: Infinite AI content is awful

#73

Post by cantor_S_drug »

We need PageRank like algorithm for "Trust / Human Content" to be applied directly to the source of such content. E.g. following all three channels are AI made. But all these content can be liked to an advanced AI version of audio based videos of Wikiarticles. If a video is providing just a summary based on established historical facts, even though it is AI based, how is it different than refering a thesaurus or dictionary? Aren't such videos making "knowledge" accessible.

FINAL Financial hours of U.S.A. just before the 1929 crash

https://www.youtube.com/watch?v=dxiSOlvKUlA&t=1008s

The Volcker Shock: When the Fed Broke the Economy to Save the Dollar (1980)

https://www.youtube.com/watch?v=cTvgL2XtHsw

How Inflation Makes the Rich Richer

https://www.youtube.com/watch?v=WDnlYQsbQ_c

Re: The trust collapse: Infinite AI content is awful

#74

Post by oblio »

heddycrow wrote: post #51

I wish we were talking about what's next versus what's increasingly here. How can infinite AI content be strictly awful if it forces us to fix issues with our trust and reward systems? Short term, sure. But infinite (also) implies long term. I wish I had a really smart game theorist friend who could help me project forward into time if for nothing other than just fun. Don't get me wrong, I'm not trying to reduce the…

> How can infinite AI content be strictly awful if it forces us to fix issues with our trust and reward systems?

You're assuming they can be fixed.

> But damned if we don't have an interesting and engaging problem on our hands right now. There's got to be some people out there who love digging in to complicated problems.

I'm sure the peasants during Holomodor also thought: "wow, what an interesting problem to solve".

Re: The trust collapse: Infinite AI content is awful

#75

Post by trylist »

arnon wrote: post #59
heddycrow wrote: post #51

I wish we were talking about what's next versus what's increasingly here. How can infinite AI content be strictly awful if it forces us to fix issues with our trust and reward systems? Short term, sure. But infinite (also) implies long term. I wish I had a really smart game theorist friend who could help me project forward into time if for nothing other than just fun. Don't get me wrong, I'm not trying to reduce the…

I suggest we go back to before and be human about things - and build trust in-person.

This is childish thinking. Whatever we do, we cannot go back to "before". Which "before"? How do we go back?

You can't regress back to a being a kid just because the problems you face as an adult are too much to handle.

However this is resolved, it will not be anything like "before". Accept that fact up front.

Re: The trust collapse: Infinite AI content is awful

#76

Post by _the_inflator »

It didn't help documentation at all. I had to work with auth0 for example and their documentation is such a bloat, that I am already prototyping with better-auth.

No structure, outdated stuff marked as "preview" from 2023/2024, wikipedia like in depth articles about everything but not for simple questions like: how to implement a backend for frontend.

You find fragments and pieces of information here and there - but no guidance at all. Settings hidden behind tabs etc.

A nightmare.

No sane developer would have done such a mess, because of time constraints and bloat. You see and experience first hand, that the few gems are from the trenches, with spelling mistakes etc.

Bloat for SEO, the mess for devs.

Re: The trust collapse: Infinite AI content is awful

#77

Post by bluGill »

bee_rider wrote: post #53
gnarlouse wrote: post #34

Earlier quoted context omitted.

We already got your money, what do we need to work for again?

Yeah. The modern software market actually seems like a total inversion of normal human bartering and trade relationships, actually… In Ye Olden Days, you go to the blacksmith, and buy some horseshoes. You expect the things to work, they are simple enough that you can do a cursory check and at least see if they are plausibly shaped, and then you put them on your horse and they either work or they don’t. Later you sell…

> In Ye Olden Days, you go to the blacksmith, and buy some horseshoes. You expect the things to work, they are simple enough that you can do a cursory check and at least see if they are plausibly shaped, and then you put them on your horse and they either work or they don’t

Nit: that is not how it worked. You took your horse to the blacksmith and he (almost always he - blacksmiths benefit from testosterone even if we ignore the rampant sexism) make shoes to fit. You knew it was good because the horse could still walk (if the blacksmith messes up that puts a nail in their flesh instead of the hoof and the horse won't walk for a few days while it heals). In 1600 he made the shoes right there for the horse, in 1800 he bought factory made horseshoes and adjusted them. Either way you never see the horseshoes until they are one the horse and your check is only that the horse can still walk.

Re: The trust collapse: Infinite AI content is awful

#78

Post by heddycrow »

arnon wrote: post #59
heddycrow wrote: post #51

I wish we were talking about what's next versus what's increasingly here. How can infinite AI content be strictly awful if it forces us to fix issues with our trust and reward systems? Short term, sure. But infinite (also) implies long term. I wish I had a really smart game theorist friend who could help me project forward into time if for nothing other than just fun. Don't get me wrong, I'm not trying to reduce the…

I suggest we go back to before and be human about things - and build trust in-person.

Dunbar's number leaps to mind. I wonder what our systems look like at large when we have cause to strengthen our 150 meaningful connections.

Would this truly be a move back? I've met people outside my social class and disposition who seem to rely quite heavily on networking this way.

Re: The trust collapse: Infinite AI content is awful

#79

Post by leviathant »

TheOtherHobbes wrote: post #23
arnon wrote: post #18

Earlier quoted context omitted.

build things """"people"""" want

People want dopamine hints, gamification, addictive distractions, and a culture of competitive perma-hustle. If they didn't, we wouldn't be having these problems. The problem isn't AI, it's how marketing has eaten everything. So everyone is always pitching, looking for a competitive advantage, "telling their story", and "building their brand." You can't "build trust" if your primary motivation is to sell stuff to you…

It boggles my mind when, despite my general avoidance of advertising online, I see the language being used. Call me old fashioned, but "viral" is a bad thing to me. "Addictive" is a bad thing. "Tricks" are bad! But this is the language being used to attract customers, and I suppose it works well enough.

Re: The trust collapse: Infinite AI content is awful

#80

Post by bluGill »

JCM9 wrote: post #70

We see business go through this cycle a lot. Some new “better cheaper” thing comes along. Everyone implements it to keep up with the Jones’s. Suddenly there’s no differentiation because everyone has it and everyone thinks it sucks. Suddenly going back to some reworked version of the old thing is the new black. One such example was call centers. In the 2000s implementing a call center in India was all the rage on cost…

Problem is how do you find real humans in the first place if you don't know them. Easy enough to walk/drive around my city and talk to them. However there are a lot of topics that the experts live elsewhere and so that won't work.
Post reply on HN