alt Hacker NewsThe other day, I remember an article was posted to HN about something, but it came from a company that provides SEO services to companies by doing something like this:
1. For a given company, analyze their target audiences and the questions they are likely to ask LLMs about.
2. For each such question, ask it to each of the major LLMs, and compute the KL divergence between the pages they want to rank for the question vs. the LLM's response.
3. Rewrite the article to minimize said KL divergence.
In effect, they're performing an iterative optimization of some sort that moves the embedding space of their article closer to the question asked to the LLM, and any embedding model or generated responses are going to prefer said responses over others.
I believe we will keep seeing more of this stuff.
Replies
There is a term for the general practice of optimizing responses called GEO - Generative Engine Optimization [0]. A cousin of SEO and equally unsavory in how trust is being eroded through info shaping. Self-discovery by individuals is the victim.
0: https://en.wikipedia.org/wiki/Generative_engine_optimization
And it's all because of ads. The incentives in an ad-funded internet are just always going to lead to this sort of thing. The most important thing is getting the user to load your page, not actually satisfying their query.
Let's hope the LLM model continues to be paying for credits, because any that move to ad revenue will become useless for real work.