Site: https://whitetower.space/
A ray dimly pierces thunderheads of the night,
Red, the crimson morning
Crawls into being
Batterd of the lightning in the night
Rivaled the sun in fractions of seconds,
Yet too extreme, the black void flooded back in.
Red sky at night, sailors’ delight,
Red sky in the morning, sailors take warning.
We’d need to train an AI that was less agreeable than current models. Current models are trained to have the “assistant” persona https://www.anthropic.com/research/assistant-axis
Hypermonetization. End up only spending effort for others.
I think this means your “smart” friends aren’t as smart as they seem. You know what, I just stopped trying and stopped going to the wrong places. Smart people don’t hang out randomly in bars because obviously most people there are dumdums. I use that time to code and get smarter. Don’t listen to average people trying to gaslight you about “social skills” if they were smart they’d know how to talk to you. Be yourself and say what you want. Average people may get triggered because they will feel like dumdums after hearing your socratic seminar and that’s your filter, that’s how you know they’re a dummy. You can try to enlighten the masses until you tire of wasting your energy playing piano for cows.
Haha conversely you can train yourself to decide it’s all a dream by saying “this is just a dream” while awake
Sometimes even I can’t open a jar. I use a C-Clamp to deform the cap by screwing into the cap diameter, and pop the air runs in and pressure lets go.
AI safety needs independent funding. What I see are companies prioritizing development, profit, or other unknown goals over safety when interests collide. The economic system is directed by funding and investment. If the goals of the investors are at odds with safe development of AI, a safety position within the established system cannot hold its own ground. AI safety needs to be self funded and self directed by its most dedicated proponents.
Who else resonates with this?
What I wonder about the OpenAI hacking episode: how did they orchestrate their agents? Anyone know? Did they use like a director agent and multiple subagents in multiple rounds? 1 round? Did they use a distributed agent hierarchy?