LLM Slop Has Only Been a Boon for Misinformation Online
Having just mentioned broader recognition that so-called 'hey hi' (AI) is just slop or mindless nonsense, citing the MSM (mainstream media), let's take a moment to consider what will be left when this bubble implodes and vanishes just like "Web3" (remember Web3?) and "blockchains".
In my own personal experience, search deteriorated. Online search engines are full of slopfarms in their indexes. Spamfarms are not a new problem to them, but slopfarms are (one similar thing that predates the LLM hype was a bunch of sites that used language translation tools to make pseudo-original and unique 'articles'). So search engines have a new challenge. Will they ever cope and/or catch up? If not, more people will gradually give up on online searches. Yes, altogether. I spoke to some people like that.
Over the past few years I tried Google a few times (typically I do not use Google) and most of the time I was astounded by how irrelevant or useless (more so upon clicking) the results were. What probably took seconds to find in the more distant past (say, a decade ago) would possibly take minutes.
The sorts of time-wasting results can be categorised/split into: spam, low-quality blurbs, social control media gossip (like Reddit or "tweets"), LLM slop, outright lunacy/misinformation, online scams (trying to sell something rogue such as fake 'money'), corporate PR disguised as 'articles', sponsored hagiography, revisionism with an SEO 'touch' etc.
The most concerning of all are the last 3, as they show how an "open" Web devolved into a plutocrats' propaganda apparatus or instrument of mass oppression 'at scale'. But I'd prefer to focus on "outright lunacy/misinformation", as that has long bothered me the most. Many are bugged a lot when their friends say utterly crazy things as if they're true, just because they saw a page or many pages online that say the same. Virtually everything can be said and spread online, more so when there's no curation or quality control. Slop makes spreading such lunacy faster and more widely. It's harder to contain misinformation and Google plays a role in spreading misinformation further. To be clear, this problem is not limited to Google [1, 2] and it cannot be tackled by media literacy alone (or sceptical approach/critical thinking skills).
LLM slop is about compensating for censorship/moderation/curation by overwhelming curators/moderators with sheer quantity. Sooner or later the moderation gets more and more automated (they try to call it "hey hi") and it gets many things wrong, including appeals (if any). Drowning out the gatekeepers with "targets" has been effective, but this tends to favour some of the worst and least ethical actors. Reckless nonsense prevails and when accountability is attempted they argue we should blame "hey hi".
There's no simple solution to all this. The very same companies that were supposed to maintain quality (again, not limited to Google with PageRank) are now actively participating in generating and spreading slop. So let's take those companies out of business. █