Jim Nielsen blogged in response to a post from Pete Millspaugh about the state of AI-generated websites in 2026.
Frankly, this all feels like a space absolutely ripe for those kinds of business referral transactions you find in other industries.
I've been thinking about this lately.
Right now, the output from Fable, Astra et al. is ostensibly an amalgamation of all source code it's slurped up - predicted text as the most likely solution for an objective. There's no ability for anyone to increase the likelihood their product or service shows up in the output via paying more (aka advertising).
At the same time, use of these agents is subsidised by the companies that offer them. They are not making money. At some point, they will have to make money. And this is where the value/utility for Users ends, and the exploitation begins, aka enshittification.
I remember when Google Search gave great results, when Facebook and Instagram genuinely kept me connected to my friends' lives, and when Twitter was a fun place to network and make connections in the industry. Every single one turned to shit in pursuit of profit.
It feels inevitable that code generating services go the same way, right? Is this as good1 as it's going to get? And how long do we have before the decline begins? 🤔
Footnotes
-
I know many of y'all would argue that agentic code output has always been, and will always be, trash. Fair point. ↩︎