AI Content Is the New Content Farm

Remember Demand Media, eHow, Associated Content, and all the other content farms that poisoned search results in the late 2000s, churning out cheap content that was mass-produced, optimized for keywords, and held to absolutely zero editorial standards until Google eventually nuked them with the Panda update in 2011?

We're doing it again.

The AI Content Gold Rush

Every SEO Twitter account is now posting about "AI content at scale," and the playbook is so obvious it hardly needs stating, pairing nicely as it does with content calendars that demand volume over quality: you feed ChatGPT a keyword list, generate thousands of articles, publish them to your site, and then presumably profit, question marks notwithstanding.

It's the same content farm model that Google supposedly killed over a decade ago, with the only meaningful difference being that the production cost dropped from $5 per article to $0.005.

The Quality Problem

AI content has a fundamental problem, one that no amount of prompt engineering or fine-tuning can solve: it can only remix what already exists.

It can't do original research, it can't have unique experiences, it can't develop genuine expertise through years of practice and failure, and it certainly can't form real opinions that emerge from actually caring about something, which means all it can do is regurgitate and rephrase existing information in slightly different configurations, which is exactly what content farms did, just slower and more expensively.

The result is an internet flooded with content that says nothing new, thousands upon thousands of articles that all read the same because they're all trained on the same data, all optimized for the same keywords, all desperately trying to rank for queries that are already answered a million times over.

Google Will Catch Up

Google killed content farms once, and they'll do it again, because right now AI content is slipping through the cracks mostly because it's new and the detection systems haven't caught up, but Google has teams working on AI detection, they have more compute power than anyone on earth, and they have every incentive to solve this problem since their entire business depends on search results being useful rather than a wall of regurgitated slop.

When the next "AI Panda" update hits, sites built on AI content will crater overnight just like the content farms did in 2011, and history doesn't repeat, but it rhymes.

The Real AI Opportunity

AI is a tool, and like any tool it can be used well or poorly: the bad use is generating content at scale to rank for keywords, treating the AI as a replacement for human thought and expertise, while the good use is speeding up research, drafting outlines, editing and improving human writing, and generating ideas to react to and argue with.

The difference is whether AI replaces human insight or augments it, and a human who uses AI to write faster and better will beat both pure AI content and pure human content, but pure AI content is just a faster content farm with better grammar.

The Sustainable Path

If you're thinking about AI content, ask yourself one simple question: Would this content exist if there were no search engines? (Understanding search intent helps answer this.)

If the answer is no, if the only reason to create it is to rank for keywords and capture traffic, then you're building a content farm regardless of how sophisticated your tools are, and it might work for a while the way all arbitrage works for a while, but it won't work forever.

The sustainable path is the same as it's always been, which is to create something genuinely valuable that helps real people with real problems: use AI to do it faster if you want, use it to research and outline and edit, but don't outsource the thinking, because the thinking is the only part that matters.

AI content at scale is just content farming with better technology. The crash is coming. Build something real instead.