However, solutions might come with the sacrifice of giving up privacy. We only have some heuristics for now, such as what existed before the AI apocalypse. AI still struggles with quality video for now, but these limitations might not last long.
This is a nice digestible summary of the ideas of The Dead Internet theory, enjoyed it as a quick read. I’ve also written about this topic, approaching it from the standpoint of examining how possible this theory is, I’d love to get your opinion on the post!
Well written article with some great references of information. At times the Inversion "marking human traffic as inauthentic and attempting to block/remove it from the network or site" seems like a reality. On X I watch as spam replies get huge amounts of views and likes while my authenticate replies get buried and sometimes even labeled as spam under "more replies" section.
You bring in some concepts I've talked about in other articles myself, such as the natural feedback loop of this phenomenon might result in less use of the internet. People move back to more real life activities. That would probably be healthy.
“Authors keep finding what appear to be AI-generated imitations and summaries of their books on Amazon. There's little they can do to rein in the rip-offs.”
This is the first item on your list that concerns me. I have hope though. Imagine that in the same way criminals now find their jobs easier, authorities will too. New cybersecurity and plagiarism-detection tools will be implemented to combat the noise, including potentially new copyright paradigms. I can’t imagine the plagiarised Amazon books will remain thereafter.
It remains however certain that the traditional browser-based internet is coming to a close. I can imagine computers inspired by the movie ‘Her’ where it’s an AGI in a monitor or phone that you interact with, not needing to touch a keyboard or mouse. The AGIs are unlikely to regurgitate plagiarised media unless the owner directly consents to it.
Yes, I hope there will be a shift in momentum as presently cybersecurity is falling behind. There aren't any good answers as of yet. AI detectors don't reliably work.
> More than 15 billion images created using text-to-image algorithms since last year. To put this in perspective, it took photographers 150 years, from the first photograph taken in 1826 until 1975, to reach the 15 billion mark.
I.e., from the *first* 150 years of photography, not the "previous 150 years" you claim in your post. Still an absurd level, but the number of photos in the *last* 150 years is still much higher, estimated in some number of *trillions* by most sources I can find from a cursory web search.
Thank you. Updated, I stated it incorrectly. The source of the linked source also mentions current numbers around 3.5 trillion, but yes the numbers are large regardless.
I no longer can easily find high quality stock images for illustrations. All the stock image sites are flooded with AI content.
It is still difficult to perceive how this will work out. Nonetheless, most apocalyptic predictions don't come true because things become intolerable before you get that far. Society eventually changes course.
However, things can still become quite miserable before that happens. Hopefully we become wiser sooner than later.
Maybe but consider this both morbid and hopeful philosophy for a moment: technology has often accelerated death and sickness, from obesity to habitat poisoning. AI accelerates perhaps its own death.
> Unlike much of the internet now, there is a human mind behind all the content created here at Mind Prison.
You can't prove that. Unfortunately.
Indeed. It is very troubling. There are some possible solutions I've mentioned in https://www.mindprison.cc/p/ai-accelerates-post-truth-civilization.
However, solutions might come with the sacrifice of giving up privacy. We only have some heuristics for now, such as what existed before the AI apocalypse. AI still struggles with quality video for now, but these limitations might not last long.
This is a nice digestible summary of the ideas of The Dead Internet theory, enjoyed it as a quick read. I’ve also written about this topic, approaching it from the standpoint of examining how possible this theory is, I’d love to get your opinion on the post!
Thank you! Do you have a link for your post?
I do! You can read the post here: https://open.substack.com/pub/cyberphilosophy/p/the-impending-solipsism-of-the-internet?r=1r9ccy&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true
Well written article with some great references of information. At times the Inversion "marking human traffic as inauthentic and attempting to block/remove it from the network or site" seems like a reality. On X I watch as spam replies get huge amounts of views and likes while my authenticate replies get buried and sometimes even labeled as spam under "more replies" section.
You bring in some concepts I've talked about in other articles myself, such as the natural feedback loop of this phenomenon might result in less use of the internet. People move back to more real life activities. That would probably be healthy.
FYI, something somewhat related that you might find interesting is a post I wrote focusing on social media - https://www.mindprison.cc/p/uniform-thought-machines
“Authors keep finding what appear to be AI-generated imitations and summaries of their books on Amazon. There's little they can do to rein in the rip-offs.”
This is the first item on your list that concerns me. I have hope though. Imagine that in the same way criminals now find their jobs easier, authorities will too. New cybersecurity and plagiarism-detection tools will be implemented to combat the noise, including potentially new copyright paradigms. I can’t imagine the plagiarised Amazon books will remain thereafter.
It remains however certain that the traditional browser-based internet is coming to a close. I can imagine computers inspired by the movie ‘Her’ where it’s an AGI in a monitor or phone that you interact with, not needing to touch a keyboard or mouse. The AGIs are unlikely to regurgitate plagiarised media unless the owner directly consents to it.
Thanks for the article.
Yes, I hope there will be a shift in momentum as presently cybersecurity is falling behind. There aren't any good answers as of yet. AI detectors don't reliably work.
FYI, I cover some of the methods attempting to counter some of these problems in my longer article https://www.mindprison.cc/p/ai-accelerates-post-truth-civilization. Search for the section "The counter measures"
From your linked source:
> More than 15 billion images created using text-to-image algorithms since last year. To put this in perspective, it took photographers 150 years, from the first photograph taken in 1826 until 1975, to reach the 15 billion mark.
I.e., from the *first* 150 years of photography, not the "previous 150 years" you claim in your post. Still an absurd level, but the number of photos in the *last* 150 years is still much higher, estimated in some number of *trillions* by most sources I can find from a cursory web search.
Thank you. Updated, I stated it incorrectly. The source of the linked source also mentions current numbers around 3.5 trillion, but yes the numbers are large regardless.
I no longer can easily find high quality stock images for illustrations. All the stock image sites are flooded with AI content.
Dead Internet may indeed save us. The crisis provides a solutiom?
It is still difficult to perceive how this will work out. Nonetheless, most apocalyptic predictions don't come true because things become intolerable before you get that far. Society eventually changes course.
However, things can still become quite miserable before that happens. Hopefully we become wiser sooner than later.
Maybe but consider this both morbid and hopeful philosophy for a moment: technology has often accelerated death and sickness, from obesity to habitat poisoning. AI accelerates perhaps its own death.
We may need a ̶B̶u̶t̶l̶e̶r̶i̶a̶n̶ ̶J̶i̶h̶a̶d̶- war.