niplav @niplav@schelling.pt

🤔

Aaahhh

[2]: https://twitter.com/jessi_cata/status/1581009760991662081

Wait. Hangon. The second name…it feels…familiar………

[1]: https://arbital.com/p/transparent_newcombs_problem/

@WomanCorn Thanks for the heads up! I would've interacted with guy 🤢

Prediction this makes: conflict theorists more often profess Knightian uncertainty about a thing

Knightian uncertainty is applicable if there is a process explicitely optimising against your belief state

@doubtshrine@stoat.zone

@tortle Our ancestors forged this network to post the least hinged excrement you could imagine. Pinnacles of poop. Cataclysms of shitposting. We honor their tradition.

@tortle

1. Solving easy problem of corrigibility (while still adhering to vNM axioms)
2. Interpretability to the level that we can extract a hand-coded algorithm from AlphaFold 2, and similar feats (maybe in total 100 bio. invested in interpretability or sth?)
3. Make GPT-4 never say *anything* violent
4. Formula for embedded diamond maximizer

Would at least make me think "ok, I should probably focus on other things"

@tortle :man_shaking_fist_at_cloud:

@amir

@vriska @sun_eater my aesthetic

It is unfortunate that positive/negative can mean "presence"/"absence" and "being for"/"being against".

It is similarly unfortunate that "phobic" can mean both "afraid of" and "being against".

I FUCKING LOVE N=1 SELF-BLINDED RCTS

@sadalsvvd I look forward to toots about astrology

@sadalsvvd nice to meet you too 🙂

@sadalsvvd helo