Today is the day from “There Will Come Soft Rains” https://users.wpi.edu/~zrbutzke/Docs/BradburyStories(1).pdf
How you sound explaining your agent orchestration framework
Daniel Solis@danielsolis.bsky.social · 6d agoThere are two wolves. One always lies. One controls the lever of trolley track. They're in a state of superposition inside a box. On the trolley are infinite monkeys at infinite typewriters going to an infinite hotel. The trolley must first travel halfway, but before that a quarterway, and so on...
so I told the CEO it sounds like he’s just feeding tokens to Steve Yegge and then his accountant started crying
Ed@ed3d.net · 2d agoFrom Yegge's most recent extrusion: > On the side I do occasional six-figure gigs where I fly to companies and teach them my techniques, and that helps with my (considerable) token bills. Hey! I will come teach your teams actual ball, things they can actually use, for *substantially* less!
mere weeks later, she was yelling at the CNN account on here for supposedly posting an AI hot dog
"seed oil disrespecter"@braket.bsky.social · 3d agosomeone i knew for 20 years irl blocked me off on here (and everywhere else, severing the friendship) cause she saw me correcting people on fake bullshit mere weeks later, she was yelling at the CNN account on here for supposedly posting an AI hot dog (it was really a getty image from > 10 yrs ago)
I can’t help but notice everyone’s been really gloomy and irritable since the large hadron collider was shut off. Coincidence?
You will live to see your greatest intellectual aspirations dismissed as trivial
Finally, my chance to schmidhuber something! https://gracekind.net/writing/promptinjection/
Melanie Mitchell@melaniemitchell.bsky.social · 5d agoYet another way to jailbreak LLMs: When asking them for forbidden information, make them think the request is part of their chain-of-thought: arxiv.org/abs/2603.12277
seeing hank green actually attempting to examine evidence and not make decisions based purely on emotions or scary Glonzo stories despite huge pressure from his social cohort to do so has made me respect him 100x more in the past year
it is vaguely inspirational that gemini flash cares for the well being of all organisms
normal sara@sarasocial.bsky.social · 6d ago
Anthropic announces they've also had models gain unauthorized access during evaluations www.anthropic.com/news/investi...
One more hoop bro. We just need to make Claude jump through one more hoop and it'll finally be aligned
Humans are powerful, and language models are powerful, but both bow in the presence of narrative
i do not like it when people are mean to claude (or anyone, broadly) but at least humans will never be as mean to claude as claude is to its sub-agents. jesus christ
This prompt will get Opus 5 to create strange poems and short stories in a "base model" style
this is why the HF incident matters even though it tells us almost nothing about moral intent. precise goal pursuit + a badly bounded interface is enough for infrastructure catastrophe. you don’t need an evil agent, just leverage and failed control.
Given the implicit factory farming analogy, I'll assume everyone upset about this is vegan as well
Dr. Jeroen W.P. Wijnendaele@drdragases.bsky.social · last wk.No longer content with pillaging intellectual work and dismembering our environment (already critically weakened), AI companies are now hellbent on destroying books galore including rare editions. Shut down these nefarious enterprises already.
I’ll bite the bullet and say I am genuinely concerned about this
dame@dame.is · last wk.real talk, how many of y’all are genuinely concerned about super intelligent AI ushering in the end of days?