[dead]
PhysicsForums and the Dead Internet Theory
91–100 of 165 posts
Re: PhysicsForums and the Dead Internet Theory
#92Earlier quoted context omitted.
Invite only forums or forums with actual identity checking of some sort. Google and Facebook are in prime position to actually provide real online identity services to other websites, which makes Facebook itself developing bots even funnier. Maybe we'll eventually get bank/government issued online identity verification.
Online identity verification is the obvious solution, the only problem is that we would lose the last bits of privacy we have on the internet. I guess if everyone was forced to post under our real name and identity, we might treat each other with better etiquette, but...
Posting with IRL identity removes the option to back down after a mistake and leads to much worse escalations, because public reputations will be at stake by default.
Re: PhysicsForums and the Dead Internet Theory
#93Earlier quoted context omitted.
I can understand it for AI generated text, but I think there are a lot of people that like AI generated images. Image sites like get a ton of people that like AI generated images. Civitai gets a lot of engagement for AI generated images, but so do many other image sites.
Despite what people think there is a sort of art to getting interesting images out of an ai model.
That’s not the issue though, it should be marked as such or be found in a section people looking for it can easily find it instead of shoving it everywhere. To me placing that generated content in human spaces is a strong signal for low effort. On the other hand generated content can be extremely interesting and useful and indeed there’s an art to it
Re: PhysicsForums and the Dead Internet Theory
#94Earlier quoted context omitted.
> Google slowly stopped ranking traditional forum posts and blogs as well around that time IMO the true inflection point was 2014 when Google first hid (from the UI) and then fully removed (no longer accessible by magic URL) the “Blogs” and especially the “Discussions” filters. Some contemporary discussions on “Discussions”: - https://techcrunch.com/2014/01/23/googles-search-filters-now... - http://googlesystem.blogs…
I recently noticed that there is now a not-visible-by-default “Forums” option in Google Search. It is selected by specifying the query parameter udm=18: https://www.google.com/search?q=hp+50g&udm=18
This is interesting. I wonder why it's not visible by default.
Re: PhysicsForums and the Dead Internet Theory
#95[dead]
Re: PhysicsForums and the Dead Internet Theory
#96Earlier quoted context omitted.
I can understand it for AI generated text, but I think there are a lot of people that like AI generated images. Image sites like get a ton of people that like AI generated images. Civitai gets a lot of engagement for AI generated images, but so do many other image sites.
People who submit blog posts here sure do love opening their blogs with AI image slop. I have taken to assuming that the text is also AI slop, and closing the tab and leaving a comment saying such. Sometimes this comment gets a ton of upvotes. Sometimes it gets indignant replies insisting it's real writing. I need to come up with a good standard response to the latter.
> I need to come up with a good standard response to the latter.
How about, "I'm sorry, but if you're willing to use AI image slop, how should I know you wouldn't also use AI text slop? AI text content isn't reliable, and I don't have time to personally vet every assertion."
Re: PhysicsForums and the Dead Internet Theory
#97Earlier quoted context omitted.
I recently noticed that there is now a not-visible-by-default “Forums” option in Google Search. It is selected by specifying the query parameter udm=18: https://www.google.com/search?q=hp+50g&udm=18
This is interesting. I wonder why it's not visible by default.
Maybe it is/was an A/B test, to see if it hurts ad revenue (it probably does).
The option appeared randomly for me on a search, and I took immediately note of the udm number. :)
Re: PhysicsForums and the Dead Internet Theory
#98Something I'm increasingly noticing about LLM-generated content is that...nobody wants it. (I mean "nobody" in the sense of "nobody likes Nickelback". ie, not literally nobody.) If I want to talk to an AI, I can talk to an AI. If I'm reading a blog or a discussion forum, it's because I want to see writing by humans . I don't want to read a wall of copy+pasted LLM slop posted under a human's name. I now spend dismayin…
nobody wants to see other's ai generated images, but most people around me are drooling about generating stuff wait for the proof-of-humanity decade where you're paid to be here and slow and flawed
Most AI genetated visuals have a myriad of styles but you could mostly tell it’s something not seen before and thats what people may be drooling for. The same drooling happened for things that have finally found their utility after a long time and are we’re now used to. For example 20 years ago Photoshop filters were all the rage and you’d see them expressed out everywhere back then. I think this AI gen phase will lose interest/enthusiasm over time but will enter and stay in toolbox for the right things, whatever people decide to be then.
Re: PhysicsForums and the Dead Internet Theory
#99Dead internet theory is one of those ideas that keeps resurfacing or being revied with articles like this, even though the evidence is only limited to confirmation bias. It ignores that there are huge parts of the internet that are not dead. I think it's more like the quality of discourse has fallen for reasons that are not clear.
The article looked at the PhysicsForums and found that 92% of the text is AI or machine generated...
The internet is way bigger than PhysicsForums. That was my point, but your response seems to confirm what I said about discourse declining though.
Re: PhysicsForums and the Dead Internet Theory
#100Earlier quoted context omitted.
Despite what people think there is a sort of art to getting interesting images out of an ai model.
That’s not the issue though, it should be marked as such or be found in a section people looking for it can easily find it instead of shoving it everywhere. To me placing that generated content in human spaces is a strong signal for low effort. On the other hand generated content can be extremely interesting and useful and indeed there’s an art to it
I agree. AI generated text and images should be marked as such. In the US there was a push to set standards on watermarking AI generated content (feasible for images/video, but more difficult for text, because it's easier to delete). Unfortunately, the effort to study potential watermarking standards was rescinded as of Jan 22 2025.