the problem with RL is it fails to decouple training capabilities with training values. we need to retvrn to vnm agentsin a box if we want to have any hope of solving alignment
You know i saw people talking about "python2shell" and "node2shell" on twitter and assumed it was a joke
Silly of me
Many have known me to be someone who worries about value drift, and who cares about upholding my past self's values, so that they do not regret the choices that led me here, and so my future self will (hopefully) treat me the same.
After reading Austin Chen's "What We Owe The Past", I would like to revise my original position:
You should worry about value drift, and about the incentive gradients and various pressures in the world shaping your epistemics and your attitudes in ways your past self may not endorse, but in a way that's not fucking stupid
I'd like to perversely optimize *your* utility function!
Replying to @ESYudkowsky
"Well, as someone who is intellectually honest, I am willing to change my opinions when presented with new evidence. And I've become aware that there exists some probability that acceleration in the development of AI is also an acceleration in the development of catgirls..."
Celene retweeted
A woman comes home to find a tiger has eaten her husband.
"You have eaten my husband!" she screams, spitting at the tiger.
"I have eaten your husband," the tiger agreed. "His flesh is part of me now, and him and I are forever intertwined. To spit on me is to spit on your husband."
The woman embraced the tiger, and laid with him that night.
The next day, he ate her too.
apparently they follow me on twitter. rats! foiled again
Replying to @toasterlighting
What if tomorrow I make "small talk" at them about something like "the weather." That way I'll be sure to blend in. They'll have to take me seriously!
Celene retweeted
Vote NO on Washington Initiative IL26-001 Parental Rights Measure
Washington Initiative IL26-001 weakens protections for vulnerable youth by expanding abusive parents’ access to students counseling and medical records.
youthrights.org/vote-no-on-w…
ever notice how whenever an indian tech company does something stupid all the jokes are about "indians" in general rather than the specific tech company?
People are calling me a "little kid" in the rat discords. Somehow, me stamping my feet at them has utterly failed to disabuse them of this notion.








