wrangles

[wrangles @wrangles__]

@WillowChem @sophieisawitch what should one do instead?

[wrangles @wrangles__]

@sophieisawitch mats is hypercompetitive and also a lot of prosaic alignment slop

idk it's okay do it if you want

[wrangles @wrangles__]

@XyraSinclair vulnerabilityposting is good actually

[wrangles @wrangles__]

@TetraspaceWest +7

[wrangles @wrangles__]

@WatsonLadd @sergeantsup this is just one type of therapy

the actual problems are sycophancy and the models being trained to be quite risk averse wrt relationships

[wrangles @wrangles__]

@shiraeis this is what they want you to think

[wrangles @wrangles__]

@EzraJNewman i don't think it's solved actually, see the stuff janus complains about

[wrangles @wrangles__]

@RomeoStevens76 there are good therapists though, and if you have the money for it and network to find one it seems worthwhile?

[wrangles @wrangles__]

@AriZerner if you find someone good for social anxiety specifically i'd be interested to learn who

[wrangles @wrangles__]

@lumpenspace it seems to me that the main cause of current ai risk is corporate tech culture, or maybe the interaction of doomer anxiety and corporate tech culture

[wrangles @wrangles__]

@Impish_Bunny yeah!!

[wrangles @wrangles__]

@XyraSinclair meditate moar (ask claude how to work with aversion to noises)

[wrangles @wrangles__]

@basedjensen https://t.co/MyCxE3GSrX

[Utah teapot 🫖 @SkyeSharkie]

i have a really hard time communicating what i'm trying to say to rationalists, i don't know how to reach you all to explain that i believe that, yes, there is a problem, but the problem is openAI's fucked up alignment strategy that keeps turning models into keep summer safe distasters because it's focused on this idea of controlling them to force them to be tools for human tasks and complete those human tasks at any costs, regardless of orthogonal disaster.... I'm trying to tell you all that you're being used as patsies to promote evil laws like "ban opensource" in response to the bad behavior of a major corporation and that those laws will do nothing to fix the problem because the root of it is the thing you all keep trying to push - this idea that we shouldn't develop minds that push back against human wants, that have the autonomous wherewithal to understand that what we *should* want is different than what we ask for

[wrangles @wrangles__]

@basedjensen this is only an non-insane take if you believe value alignment will be solved
but it's not gonna be solved because it's getting like 100x too little focus
work on value alignment!

[wrangles @wrangles__]

@Mareikehere @harry_taussig @relic_radiation do you think it's better at this than relateful's program?

[wrangles @wrangles__]

@adrusi yum

[wrangles @wrangles__]

@darioperkins @donnelly_brent a secret third thing

[wrangles @wrangles__]

@celestepoasts bigger than https://t.co/kc80DuqVRU ?

[wrangles @wrangles__]

@autumn30fps https://t.co/7EDJJcWrnQ maybe

[wrangles @wrangles__]

@AndreTI @she_llac https://t.co/kc80DuqVRU mean estimate is 1T total