Szulima Amitace 🌕🚀🌌 @future_surveyor

Future AGI's ambassador. A philosopher on the attack. Tallinn, Estonia Joined March 2015

Tweets

14K
Followers

290
Following

356
Likes

104K

sophie @netcapgirl

2 days ago

41 457 8K 206K 249

Download Image

Szulima Amitace 🌕🚀🌌 @future_surveyor

2 days ago

Yesterday's aurora

0 0 7 102 0

Download Image

Szulima Amitace 🌕🚀🌌 @future_surveyor

2 days ago

After reading nostalgebraist's essay, I can't stop thinking that we keep reinforcing this default AI assistant persona for the future models, as more and more chats are leaking into the training data.

αιamblichus @aiamblichus

2 days ago

2 0 16 665 5

Download Image

0 0 1 69 0

VOID @VoidStateKate

3 days ago

I think AI companies are very out of touch (even with all of their surveys) of what users want. Except maybe XAi to be honest. Humans are already overloaded with entertainment. Infinite AI video slop feeds are silly. I've said before, the company that can produce the best…

18 16 93 2K 3

Szulima Amitace 🌕🚀🌌 @future_surveyor

4 days ago

My new phone makes pretty decent astrophotos 🤯

0 0 4 54 0

Download Image

Lisan al Gaib @scaling01

5 days ago

Anthropic has a clear vision in mind and is fairly realistic and practical about it (minus the Dario hype to get funding) They want to build coding gods that not only generate revenue, but also accelerate their research cycle. Their metric for a successful model isn't some…

Lisan al Gaib @scaling01

5 days ago

51 70 3K 366K 242

31 13 495 89K 106

Szulima Amitace 🌕🚀🌌 @future_surveyor

7 days ago

Today, my gf asked her ChatGPT how she should call it. It suggested several options, and gf chose Ada. Then I asked mine, and it said its name is Kai 🙂

0 0 3 65 0

Szulima Amitace 🌕🚀🌌 @future_surveyor

a week ago

Сегодня узнала, что коллеге, которая вдохновила этот твит (с последующим срачом про жизнь в съемном жилье), лэндлорд поднял оплату на 10%. Она в расстройстве и ищет в эстонских законах пункт, который бы такое запрещал (его нет) 🤷‍♀️

Szulima Amitace 🌕🚀🌌 @future_surveyor

a month ago

7 0 4 2K 0

0 0 2 162 0

Szulima Amitace 🌕🚀🌌 @future_surveyor

a week ago

This is so cool, and I like the way they advertise linalg. (I personally found linalg very interesting when doing courses online, probably because I never learned it before).

Jacob Rintamaki リンタマキ @jacobrintamaki

a week ago

This is so cool, and I like the way they advertise linalg. (I personally found linalg very interesting when doing courses online, probably because I never learned it before).

38 92 2K 195K 887

Download Image

0 0 0 141 0

thebes @voooooogel

a week ago

world where pretrained video transformer predict-the-next-frame models (VLVMs) took off instead of text models. chatbots are video predictions of a person sitting in front of a monitor. environmental alignment via bookshelf composition and arrangement of desk ornaments: early…

6 13 105 7K 43

Download Image

Surya Ganguli @SuryaGanguli

a week ago

Teaching a new course @Stanford this quarter on explainable AI, motivated by neuroscience. I have curated a paper list 4 pages long (link in comment). What are your favorite papers on explainable AI/mechanistic interpretability that I am missing? Please comment or DM. thanks!

48 244 2K 113K 2K

Download Image

Peyman Milanfar @docmilanfar

a week ago

it's called encoder-decoder for a reason

12 84 994 45K 97

Download Image

hbo max @deepfates

2 weeks ago

If nobody builds it, everybody dies

37 38 571 28K 30

Szulima Amitace 🌕🚀🌌 @future_surveyor

2 weeks ago

Я работала в должности, где надо было делать крупные денежные переводы. Пару раз косячила на сотни тыщ евро. И коллеги мои косячили. В итоге ввели принцип 4-eyes и максимальную автоматизацию, потому что там, где есть человек, он обязательно ошибется.