> or a spammer where your incentives have probably increased.
Slight pushback on this. The web has been spammed with subpar tutorials for ages now. The kind of medium "articles" that are nothing more than "getting started" steps + slop that got popular circa 2017-2019 is imo worse than the listy-boldy-emojy-filled articles that the LLMs come up with. So nothing gained, nothing lost imo. You still have to learn how to skim and get signals quickly.
I'd actually argue that now it's easier to winnow the slop. I can point my cc running in a devcontainer to a "tutorial" or lib / git repo and say something like "implement this as an example covering x and y, success condition is this and that, I want it to work like this, etc.", and come back and see if it works. It's like a litmus test of a tutorial/approach/repo. Can my cc understand it? Then it'll be worth my time looking into it. If it can't, well, find a different one.
I think we're seeing the "low hanging fruit" of slop right now, and there's an overcorrection of attitude against "AI". But I also see that I get more and more workflows working for me, more or less tailored, more or less adapted for me and my uses. That's cool. And it's powered by the same underlying tech.
The thing is, what is the actual point of this approach? Is it for leaning? I strongly believe there’s no learning without inmersion and practice. Is it for automation? The whole idea of automation is to not think about the thing again unless there’s a catastrophic error, it’s not about babysitting a machine. Is it about judgment? Judgment is something you hone by experiencing stuff then deciding whether it’s bad or not. It’s not something you delegate lightly.
The problem isn't that AI slop is doing something new. Phishing, blogspam, time wasting PRs, website scraping, etc have all existed before.
The problem is that AI makes all of that far, far easier.
Even using tooling to filter articles doesn't scale as slop grows to be a larger and larger percentage of content, and it means I'm going to have to consider prompt injections and running arbitrary code. All of this is a race to the bottom of suck.
> The problem isn't that AI slop is doing something new. Phishing, blogspam, time wasting PRs, website scraping, etc have all existed before. The problem is that AI makes all of that far, far easier.
The term I like is that AI has _industrialised_ those behaviours. While native hunted buffalo, it wasn't destructive until it was industrialised [1] it that it became truly destructive.
The difference is that the cost of slop has decreased by orders of magnitude. What happens when only 1 in 10,000 of those tutorials you can find is any good, from someone actually qualified to write it?
One instance of definite benefit of AI is AI summary web search. Searching for answers to simple questions and not having to cut though SEO slop is such an improvement
The summary is often incorrect in at least some subtle details, which is invisible to a lot of people who do not understand LLM limitations.
Now, we can argue that a typical SEO-optimized garbage article is not better, but I feel like the trust score for them was lower on average from a typical person.
Marketing departments are already speaking of GEO - generative engine optimization. When a user asks an AI for the best X, you want it to say your X is the best, and you'll do whatever it takes to achieve that.
Google was unable or unwilling to fight people gaming their SEO to float garbage and blogspam to the top results, waay before these more specific policy change events that have been reported w.r.t intentionally making search worse.
granted it's not up to courtroom standards, this post linked by another commenter in the chain does paint the picture pretty well of an internal struggle between Search and Ads inside Google as a company, where there was a decision to promote user-negative changes to Search as a way to increase the total number of searches performed, thereby increasing the number of ads that can be shown. This happened during 2019.
I don't understand this position, do you have direct evidence that Google actively made search worse? Before I'm misunderstood I do want to clarify that IMO, the end user experience for web searching on Google is much worse in 2025 than it was in say 2000. But, the web was also much much smaller, less commercial and the SNR was much better in general.
Sure, web search companies moved away from direct keyword matching to much more complex "semantics-adjacent" matching algorithms. But we don't have the counterfactual keyword-based Google search algorithm from 2000 on data from 2025 to claim that it's just search getting worse, or the problem simply getting much harder over time and Google failing to keep up with it.
In light of that, I'm much more inclined to believe that it's SEO spam becoming an industry that killed web search instead of companies "nerfing their own search engines".
>These emails — which I encourage you to look up — tell a dramatic story about how Google’s finance and advertising teams, led by Raghavan with the blessing of CEO Sundar Pichai, actively worked to make Google worse to make the company more money. This is what I mean when I talk about the Rot Economy — the illogical, product-destroying mindset that turns the products you love into torturous, frustrating quasi-tools that require you to fight the company’s intentions to get the service you want.
Of course, it's hard to "objectively" prove that they literally made search worse, but it's clear they were fine with stagnating in order to maximize ad revenue.
I see it as the same way Tinder works if you want the mentality. There's a point where being "optimal" hurts your bottom line, so you don't desire achieving a perfect algorithm. Meanwhile, it can be so bad for Google that directly searching for a blog title at times can leave me unsuccessful.
> I see it as the same way Tinder works if you want the mentality. There's a point where being "optimal" hurts your bottom line, so you don't desire achieving a perfect algorithm
Yes, in the case of Google:
- They make more money from ads if the organic results are not as good (especially if it's not clear they're add)
- They get more impressions if you don't find the answer at the first search and have to try a different query
This is entirely because "we" insist on search being free. This means Google needs to find other ways to pay for it, which creates a different set of incentives.
If we somehow paid directly for search, then Google's incentives would be to make search good so that we'd be happy customers and come back again, rather than find devious ways to show us more ads.
Most people put up with the current search experience because they'd rather have "free" than "good" and we see this attitude in all sorts of other markets as well, where we pay for cheap products that fail over and over rather than paying once (but more) for something good, or we trade our personal information and privacy for a discount.
> In light of that, I'm much more inclined to believe that it's SEO spam becoming an industry that killed web search instead of companies "nerfing their own search engines".
"SEO" is not some magic, it is "compliance with ranking rules of the search engine". Google wanted to make their lives easier, implemented heuristics ranking slop higher, resulting in two things happening simultaneously: information to slop ratio decreasing AND information getting buried deeper and deeper within SRPs.
> do you have direct evidence that Google actively made search worse?
https://support.google.com/google-ads/answer/10286719?hl=en-...
Google is literally rewriting the queries. Not only results with better potential for ads outrank more organic results, it is impossible to instruct the search engine to not show you storefronts even if you tried.
Before Google the web was already full of SEO slop. Except "SEO" just meant "put a list of popular keywords in a block hidden with CSS". That was the time of Altavista and other search engines.
So the time you're talking about is a window when Google existed, but before they gave up on fighting spam.
Hard disagree. AI summaries are useless for the same reason AI summaries from Google and DDG are useless: it's almost always missing the context. The AI page summaries typically take the form of "here's the type of message that the author of this page is trying to convey" instead of "here's what the page actually says". Just give me the fucking contents. If I wanted AI slop I'd ask my fucking doorknob.
I think you have some of your wires crossed, asking Google for "here's the type of message that the author of this page is trying to convey" is not what most people think is a simple question (also asking Google to reprint copyrighted material us also a non starter). Asking Google "what is the flag for persevering Metadata using scp" and getting the flag name instead of a SEO article with the a misleading title go on about so third party program that you can download that does exactly that and never actually tell you the answer is ridiculous and I am happy AI has help reduce the click bait
"here's the type of message that the author of this page is trying to convey" is not what most people think is a simple question
It's also not the question I asked. I'm literally trying to parse out what question was asked. That's what makes AI slop so infuriating: it's entirely orthogonal to the information I'm after.
Asking Google "what is the flag for persevering Metadata using scp" and getting the flag
name instead of a SEO article with the a misleading title go on about so third party program
that you can download that does exactly that and never actually tell you the answer is
ridiculous and I am happy AI has help reduce the click bait
Except that the AI slop Google and Microsoft and DDG use for summaries masks whether or not a result is SEO nonsense. Instead of using excerpts of the page the AI summary simply suggests that the SEO garbage is answering the question you asked. These bullshit AI summaries make it infinitely harder to parse out what's actually useful. I suppose that's the goal though. Hide that most of the results are low quality and force you to click through to more pages (ad views) to find something relevant. AI slop changes the summaries from "garbage in, garbage out" to simply "garbage out".
At least with Google, it quotes the pages where it gets the information from. Also, I think you are definitely underplaying the fact that it answers the question in one sentence,as wellas the whole ask a question get a compact answer. I am going to need a concrete example, because in my experience, the AI summary has never even required me to verify the source except out of curiosity, much less click on any search results.
Sadly these aren't monkeys. These are more like termites eating at any and all wood they can find. They'll eat at the foundation and move to the next trend to eat at.
Spam by its nature is low effort, low yields anyway. They don't particularly care about making scraps since their pipeline is nearly automated.
Slight pushback on this. The web has been spammed with subpar tutorials for ages now. The kind of medium "articles" that are nothing more than "getting started" steps + slop that got popular circa 2017-2019 is imo worse than the listy-boldy-emojy-filled articles that the LLMs come up with. So nothing gained, nothing lost imo. You still have to learn how to skim and get signals quickly.
I'd actually argue that now it's easier to winnow the slop. I can point my cc running in a devcontainer to a "tutorial" or lib / git repo and say something like "implement this as an example covering x and y, success condition is this and that, I want it to work like this, etc.", and come back and see if it works. It's like a litmus test of a tutorial/approach/repo. Can my cc understand it? Then it'll be worth my time looking into it. If it can't, well, find a different one.
I think we're seeing the "low hanging fruit" of slop right now, and there's an overcorrection of attitude against "AI". But I also see that I get more and more workflows working for me, more or less tailored, more or less adapted for me and my uses. That's cool. And it's powered by the same underlying tech.