In my bleaker moments, it really does scare me how wildly far our culture is from proper regulation to protect people against the abuses of LLM-based technology.
Yes, I'm using the A word. I find the idea of being forced to engage with genAI to be abusive.
You want me to read generated text or watch generated video and I don't have a “choice” in the matter (because it’s part of the some societal process like applying for a job or getting healthcare)? I consider that a form of harassment.
I have a horrible sneaking feeling that, until if/when there’s a crash, an important skill in this brave new world will be working out which foundation model is being used and then tweaking your CV appropriately
Because all these products _will_ have repeatable and abusable eccentricities