Jason LefkowitzThere are lots of ways that AI is eroding the intellectual commons, but a subtle one is that now the discussion around every single essay and blog post is immediately dominated by a debate over whether or not it was written with AI
historic drystone sheep dyke@jalefkowit the exact inverse problem is that i saw this piece posted on metafilter a couple of days ago and i felt crazy because nobody else was talking about the fact that the illustrations are almost definitely ai, which feels egregious considering the topic https://bachmanrachel.substack.com/p/what-children-actually-want-from
@hannah Yeah, I would say it's all of a piece; you can't engage with the substance of a work anymore without first establishing how much of it is from the author's own hand and how much is AI, and there's no independent way to do that, so you end up squinting at every line, every illustration, every chart, asking yourself, can I trust this? Is this real?
It's exhausting, which is why it makes me fear for the future of thought. I find myself turning away from things just because I don't want to have to be the Em Dash Police
JJ Litke@jalefkowit @hannah "Em Dash Police" <— another frustrating bit, because I fucking love em dashes, and now I feel like I need to edit them out of my writing entirely.
@jalefkowit yeah it feels like a gresham's law thing where in a few years the open internet will just be 99.99% llm spam like what happened to usenet, and we'll all have to go back to small trusted sites or private group chats. oh well
Tyler SmithI read this described as breaking a social contract. Pre-AI, the writer always put more time into writing a piece than the reader would spend reading it. In effect, they were giving you X hours(days/months) of their work, hoping to earn Y minutes of your attention.
AI has inverted this. The writer now demands Y minutes of our attention in exchange for X seconds of their 'effort'. It's anti-social narcism: my half-baked idea is worth your careful consideration.
And of course there's the knock-on effect you describe, in that we now have to interrogate every piece of writing we encounter to determine if it's a good faith expression of someone's thoughts, or just some fleeting thought inflated to a grotesque imitation of human communication.
mathew@hannah @jalefkowit It’s on Substack, so I wouldn’t have expected any strong ethical principles.