Less is more, but it’s already all ‘too much’: The rise of digital disengagement
Photo: Timon Studler
A new Europol report has predicted that as much as 90% of all content on the internet could be ‘synthetically generated’ (made or modified by AI) by 2026. That is a two-and-a-half-year window and closing, with the end result being only a small minority of online content having been conceived of, much less created by, actual humans.
The presence of bots is nothing new; indeed, Elon Musk made a war on bots a part of his Twitter takeover. However, a few bots here and there generating chaos in the comments section or spamming posts that everyone knows are fake is one thing; the internet becoming a sea of artificially-generated content is another, with very few original fish.
There is another inherent problem with this from an AI standpoint. As AI content permeates the existing web, the data that AI is trained on – which is taken from the same web – is increasingly going to be influenced by, well, itself. Cue a cyclical creative decline of proliferating iteration in an effective vacuum. Perhaps artists should not be so frightened, after all; humans may be able to discern the good from the bad through their capacity for enjoyment of it, but AI has no such ear, and its already somewhat garbled output can only get worse when it learns mainly from itself.
The near-term impacts, however, are already being felt. Twitter and Reddit have made moves to limit the scraping that can be done on their sites – not necessarily to prevent developers from making third-party apps or simply for fear of bots diluting their content value (both somewhat true), but more so because their sites are prime ground for the data collection that feeds into the AI with no costs associated. Much like visual artists, musicians, and writers, social platforms are seeing rights (and revenue)-free scraping benefitting the AI companies' potential margins, while having no returns of their own to show for it. A bad time, considering the economic downturn is causing shareholders to come asking for actual revenue returns, rather than simply user growth.
Fandom and identity For fans, entertainment goes beyond enjoyment of content
Across entertainment, fandom is used to judge, predict, and generate successful commissioning and distribution strategies. Fandom indicates reliable audiences who will spend money on propositions and...Find out more…
All this is serving to quell the until-now thriving digital culture of the ‘global town square’; limitations that keep out the bots also keep people out as well (there has been much unrest over the new Tweet limits during the current airing of Love Island), and the feedback loop of rising amounts of AI content are quelling the enthusiasm users have for the platforms as social networks. This is because the content on them can no longer be trusted to be from the real people they want to interact with. Similarly, free-to-use social apps do not earn lots of money, but ads do – and those are better suited to fit between short-form videos on an infinite scroll than in between thought-provoking conversation topics online. Yet, this leaves a massive gap in the market for actual social networking platforms.
Ironically, entertainment and fandom-focused hubs are becoming prime places for connection (a finding backed up by our latest media report). Gaming offers socialisation in an engaged, active context with like-minded people; and fan communities thrive on the likes of TikTok (which is itself facing bans, or at least increased regulation). An opening in the market is appearing in this space, with the likes of ArtistVerified hoping to move in and offer a fan-focused community for building connections between artists and audiences.
However, the trajectory is becoming clear. Social platforms are increasingly becoming interactive content platforms. Socialisation is happening everywhere – in real life, on private messaging, and in gaming sessions. And users are increasingly being faced with digital social spaces filled with unreliable AI-generated content, ads, and the most intrusive thoughts from the most chronically online denizens of humanity – resulting in the potential for digital disengagement. We are already seeing increased vinyl sales; film cameras are back on the rise, live events are increasingly cut-through moments made consciously irreplicable in the digital world in order to stand out, and social media ‘cleanses’ are a growing trend. Not to mention, of course, a global economic crisis is not a great environment for anything new and pricey to take off. Thus, VR may have had its moment at a bad time; AI might spell a downturn for the perceived value of the digital world altogether, and digital entertainment is continuing to oversaturate itself. And, of course, Mark Zuckerberg and Elon Musk are training for a cage wrestling match, because why not – even the heads of the biggest incumbent players in the space, which built the very cultural infrastructure of web 2.0, are taking their professional trajectories that seriously.
AI has accelerated everything, not least among it our collective fascination with the internet. The hype cycle is collapsing beneath the weight of itself. The rise of digital disengagement may be a long-term trend… but everything is faster in the world of web 3.0, so it may not be nearly as far on the horizon as one might hope (or worry).