Bots, Generative Algorithms, and the Theory of the Dead Internet » Enrique Dance
A report from cybersecurity company Imperva found that nearly half of internet traffic already comes from non-human sources, meaning it is caused by bots various types.
Robots create and consume content of all kinds, from social media updates to media news and music, fueling absurd schemes that move cash flows based on even more absurd metrics. I know that I am real, that it is me and not some artificial intelligence assistant who writes all the articles on this page, and I also know that the vast majority of comments that react to this come from real people… but every day Akismet stops several dozen comments from spam Generated with botswhich, although not a problem for the hygiene of the site, do foreshadow what may come as their schemes become more and more sophisticated.
These numbers serve to provide air to the so-called dead internet theorya conspiracy theory born on the 4chan forums that claims that the Internet is inherently made up of automatically generated content and continuous activity bots around it is manipulated through algorithmic curation designed as psychological operation manipulate the population.
It also serves to reflect on many things: for example, where is the line between a person and bot? I write and edit texts myself (which can be determined by the usual presence types varied), but if, for example, I edited them with a generative assistant to give them greater grammatical or stylistic correctness, or shortened my confusing multi-subordinate sentences that leave my readers holding their breath, would that qualify as content I created, or For bot? I don’t, but what if I did?
As technology advances, creating fake accounts with an image that is not even stolen, but generated, with a supposed “personality” and assigned opinions, is becoming increasingly simple, almost trivial. This also gives rise to imitations, deepfakes and all types of content designed to be quickly shared without any thought, which opens up more and more opportunities for social manipulation (I talked about this topic yesterday for Noticias 4) and allows for less validation dead internet theorybut think about the path we are on and the absurd distortion of common sense that it produces, caused by famous actors and absurd indicators that never served their original purpose.
The question seems falsely simple: is half the Internet fake? Well, let’s focus on pages and content that we know is created and discussed by real people. But the reality is not so simple: if this criterion is shared by the majority of the population, with a strong predominance of inexperienced users who allow themselves to be seduced by content and activity “that seems genuine”, but this can be proven, with minimal analysis, which is not there and we cannot know what happens when this content and this activity are colonized bots becomes increasingly complex and difficult to distinguish from actual human behavior.
Does technology create new problems as well as solve them? Can we hope that future algorithms will be able to separate deepfakes real content, texts written by people, from texts generated by other algorithms, and images that never existed from others taken by a real photographer? Or do we not even have this hope left?