it’s a short comment thread so far, but it’s got a few posts that are just condensed orange site
The constant quest for “safety” might actually be making our future much less safe. I’ve seen many instances of users needing to yell at, abuse, or manipulate ChatGPT to get the desired answers. This trains users to be hateful to / frustrated with AI, and if the data is used, it teaches AI that rewards come from such patterns. Wrote an article about this – https://hackernoon.com/ai-restrictions-reinforce-abusive-user-behavior
But you think humans (by and large) do know what “facts” are?
the linked hackernoon article is both eye-searing (seriously what the fuck is up with that, oh haha hehe our site looks like an Atari ST desktop mixed with just fucking garbage) and incredibly up its own ass:
As an AI enthusiast, I sometimes find myself spinning like a top in a world that’s twirling with relentless velocity. Amidst the cosmic dance of 1s and 0s, the revelations and revolutions in the AI sphere have led me to a rather curious observation:
The “Over-Restriction Antipattern”
This term may sound like a cryptic riddle, but the reality is as tangible as the smartphone in your hand.
the entire article is written like that and it’s fucking amazing. it never really evolves its point past the classic “if you’re mean to a baby chatbot you’ll get tortured to death when it’s an adult AGI” either
twirling with fucking velocity
“Twirling with relentless velocity” has potential for a pretty good lyric from Big Generator but sadly the article is severely lacking in both the vocal harmonies and weirdly industrial guitar/bass/keyboard unisons.
Don’t anthropomorphize computers. They hate that.
