| ▲ | vhiremath4 4 hours ago |
| From most people I talk to, there is an understanding that the overwhelm of AI content, bot comments, etc. has degraded the quality of community spaces (especially HN but also including Reddit, X/Twitter, etc.). I hope we can figure out a way to get more human-curated content soon and make the economics of amplifying it make sense (obviously it does not in this current transition chapter). |
|
| ▲ | ungovernableCat 3 hours ago | parent | next [-] |
| I'd guess more and more valuable discussion will just shift to gated invite only groups. Private discord servers, signal, telegram etc etc
I don't see another feasible way of protecting against the spam. What's more strange to me is why these content places like Twitter and Reddit don't see this as a danger to themselves? Why would someone want to advertise in a place that's 90% bots? Reddit for example made it harder to check astroturf bots by blocking viewing comment and post history and giving every user default AdjectiveNounNumber style names. I get why it works short term but I've completely stopped using it because I just found myself facing fake botted engagement everywhere. |
| |
| ▲ | pixeljam 27 minutes ago | parent | next [-] | | > I'd guess more and more valuable discussion will just shift to gated invite only groups. Private discord servers, signal, telegram etc etc I don't see another feasible way of protecting against the spam. This is sadly the direction I think we are heading towards. I will lament the loss of the open internet. > Reddit for example made it harder to check astroturf bots by blocking viewing comment and post history and giving every user default AdjectiveNounNumber style names They did this just weeks before their IPO in what I assume was an effort to decieve shareholders from discovering just how did the platform really is. The amount of "people" with that setting on is shocking to me. On a site where people didn't even set a profile picture 9/10 times, I cannot believe that like 7/10 users are choosing to hide their history.
Reddit is playing an illusion game fueled by bots to convince users and shareholders the platform is alive when it has began dying years ago. It's also quite shocking to me that they do this and then choose to make their tagline "Where Real People Talk": https://apps.apple.com/nz/app/reddit-where-real-people-talk/... | |
| ▲ | vhiremath4 3 hours ago | parent | prev [-] | | This has also perplexed me (advertising-to-bot-relationship), but I'm guessing the cost of combating these bots is going up (as complexity goes up and token cost goes down - whether subsidized or not). I hate the idea that the future is only gated invite-only groups though. At the very least, some tag of "BOT" and then the ability to hide all bot content (unless it's highly upvoted by humans) would be great. | | |
| ▲ | ungovernableCat 3 hours ago | parent [-] | | Any attempt to seriously combat bots would most likely impact real user engagement numbers so it is indeed a tricky business problem to solve. But I can't see the Reddit comment change as anything other than a cynical "astroturfing as a service" to advertisers. Doesn't seem to be working out for them though given their stock performance this year. |
|
|
|
| ▲ | xiaoyu2006 3 hours ago | parent | prev | next [-] |
| I propose we sign each other's PGP keys as a honor code system of writing of one's own words. Idk how this will turn out... Complete distrust or tribalism or actually useful? |
| |
| ▲ | ezekg 3 hours ago | parent | next [-] | | The problem is that it's too easy to copy/paste a rebuttal written by an LLM as if you wrote it, without actually understanding it. Simply signing such a faux thought does nothing. | |
| ▲ | foobarian 3 hours ago | parent | prev | next [-] | | Maybe we could have meetups to do this signing to make things more efficient | | |
| ▲ | esafak 3 hours ago | parent [-] | | Organizing physical meetings with HN's global community is an efficient solution?? | | |
| ▲ | Timwi 2 hours ago | parent [-] | | Maybe an online meetup on Zoom is just about still viable, but we need to hurry up before the AI face filters learn how to deal with fingers being waved in front of your face. |
|
| |
| ▲ | p-e-w 3 hours ago | parent | prev [-] | | I imagine the result would be a mixture of elitism, community fragmentation, and a witch hunt that never stops. Like Stack Overflow meets The Crucible. |
|
|
| ▲ | Rapzid 2 hours ago | parent | prev [-] |
| Github projects too. Any popular project drowns human voices out in an ocean of AI vomit. Issues, comments, replies, PRs.. All from the slop firehose. Makes trying to interact with them miserable. |