61–70 of 369 posts
AI products will be speed running the last few decades of tech - make a compelling product, then add ads based on the prompt, then store prompts from each user and build psychological profiles, and finally manipulate AI output to maximize user susceptibility to part with their hard earned money. Bonus points if they can get the user addicted to their product.
I think a difference here is that a person who is discerning enough to use an LLM product to read more information on a given subject is probably going to be turned off by the horrible UX of ads. YouTube and Facebook and Google can be plastered with ads on every page because the lowest common denominator of user wants to use those products. One of ChatGPT's main draws was the clean no bullshit, Apple-like UI and I'm…
Once one us ads others follow. Not up to the user.
Even Netflix has ads …
Earlier quoted context omitted.
What could make a streaming service "decent" ?
Spotify but for video media.
Not precisely what I had in mind from the term "decent".
ps. I know that for most users of Spotify, the above description of the service doesn't correspond to their reality. But it is an accurate description of the service, nevertheless.
No surprise here, but what a bummer that this is "The Way™" Just let me pay you $3-5/ month to never see an ad again. It's worth it to me since the service is actually decent.
You mean Perplexity Pro? That thing they tried and found that no one was willing to pay for cuz users say they want paid options but then jump boat to the free things always? https://www.perplexity.ai/hub/faq/what-is-perplexity-pro
Google got off the ground because they had magic sauce that made their product noticeably better. LLM-based search engines don’t have that, especially when they’re using an LLM built elsewhere.
AI products will be speed running the last few decades of tech - make a compelling product, then add ads based on the prompt, then store prompts from each user and build psychological profiles, and finally manipulate AI output to maximize user susceptibility to part with their hard earned money. Bonus points if they can get the user addicted to their product.
Without ads the world would basically stop, they drive the businesses, each and every one.
It’s a necessary evil that will always be in most popular products, especially considering most people want to get all value for free.
AI products will be speed running the last few decades of tech - make a compelling product, then add ads based on the prompt, then store prompts from each user and build psychological profiles, and finally manipulate AI output to maximize user susceptibility to part with their hard earned money. Bonus points if they can get the user addicted to their product.
Earlier quoted context omitted.
How do you block ads if they are embedded in the training weights?
I interpreted the suggestion as the reader's own AI would rewrite the content to remove advertising messaging/bias, for example.
There's no defense against a future in which an AI response has undisclosed purchased bias. Regulation is impotent.