A year ago, author and journalist Cory Doctorow highlighted the ways in which a platform dies: "First, they are good to their users; then they abuse their users to make things better for their business customers; finally, they abuse those business customers to claw back all the value for themselves. Then, they die."
It is a process he called "enshitification."
He highlighted both Amazon and Google Search, saying of the latter: "Google's results are an increasingly useless morass of self-preferencing links to its own products, ads for products that aren't good enough to float to the top of the list on its own and parasitic SEO junk piggybacking on the former."
Just a few months later, AI researcher and cognitive scientist Gary Marcus borrowed the phrase to highlight a newer source of the Search cesspool he was witnessing: The "Large Language Model (LLM)-driven enshitification of the internet."
Related: AI tax fraud: Why it's so dangerous and how to protect yourself from it
Marcus said in February of 2023 that LLMs like ChatGPT can be used to produce fake reviews and synthetic websites — designed for Search Engine Optimization (SEO) — at scale.
While such models "aren't very good at writing error-free prose," Marcus said that people running content farms "won't care about the errors."
Instances of this are rapidly growing; 404 Media in January discovered a series of AI tools designed to publish synthetic iterations of already-published articles at scale.
One called SpinRewriter "takes a single article and turns it into dozens of 100% unique, human-quality articles with ZERO machine-generated footprint. This lets you rank higher, and for more profitable keywords."
Another, Byword, advertises itself with this simple phrase: "Generate high quality, AI-written articles at scale."
"Search itself seems likely to face a major growing threat from circles of fake websites that are designed to sell ads and mutually reinforce each other’s positions in Search Engine Optimization," Marcus said in February.
"Cesspools of automatically-generated fake websites, rather than ChatGPT search, may ultimately come to be the single biggest threat that Google ever faces," he added. "After all, if users are left sifting through sewers full of useless misinformation, the value of search would go to zero — potentially killing the company."
Related: Popular social media platform to sell user data to the company behind ChatGPT
Google's new approach
Google said Tuesday that it is changing its approach to ensure the quality of Search.
The new tactics are centered around three different types of abuses: Scaled content abuse, site reputation abuse and expired domain abuse.
Google said in a blog post that, though it has had policies in place against the use of automation to generate low-quality content at scale, it is becoming increasingly difficult to identify the role of AI in the process of content creation. Because of this, Google said that it is focusing on addressing and reducing the abuse itself — low-quality content at scale tailored to boost search rankings — rather than the possibly automated method behind that abuse.
The update, Google said, involves the refining of its core ranking systems.
More deep dives on AI:
- New platform seeks to prevent Big Tech from stealing art
- Marc Benioff and Sam Altman at odds over core values of tech companies
- Senate Judiciary Committee seeks to build new framework to rein in Big Tech
Site reputation abuse, according to Google, involves the use of low-quality third-party content on otherwise respectable sites. The owners of those sites will have two months to make changes before they're hit with Google's new ranking methodology.
Expired domain abuse involves the purchase of an expired — but once respectable — domain as a shell to pump out low-quality articles designed with SEO optimization in mind.
Google said that such domains are now considered spam.
Google expects the update to reduce low-quality Search results by 40%, at the same time sending "more traffic to helpful and high-quality sites."
Contact Ian with tips and AI stories via email, ian.krietzberg@thearenagroup.net, or Signal 732-804-1223.
Related: Gary Marcus says Elon Musk 'has a point' with latest lawsuit against Sam Altman, OpenAI
