RSS

Michael Tontchev

Karma: 176

GPT-4 can catch sub­tle cross-lan­guage trans­la­tion mistakes

Michael TontchevJul 27, 2023, 1:39 AM
7 points
1 comment1 min readLW link

[Question] Do you speed up ca­pa­bil­ities when you do AI in­te­gra­tions and con­sume over­hangs?

Michael TontchevJul 20, 2023, 6:40 AM
6 points
1 comment1 min readLW link

[Question] Links to dis­cus­sions on so­cial equil­ibrium and hu­man value af­ter (al­igned) su­per-AI?

Michael TontchevJul 8, 2023, 1:01 AM
7 points
3 comments1 min readLW link

Outreach suc­cess: In­tro to AI risk that has been successful

Michael TontchevJun 1, 2023, 11:12 PM
83 points
8 comments74 min readLW link
(medium.com)

A rough model for P(AI doom)

Michael TontchevMay 31, 2023, 8:58 AM
0 points
1 comment2 min readLW link

Align­ment solu­tions for weak AI don’t (nec­es­sar­ily) scale to strong AI

Michael TontchevMay 25, 2023, 8:26 AM
6 points
0 comments5 min readLW link

Unal­igned sta­ble loops emerge at scale

Michael TontchevApr 6, 2023, 2:15 AM
9 points
8 comments4 min readLW link

ChatGPT’s “fuzzy al­ign­ment” isn’t ev­i­dence of AGI al­ign­ment: the ba­nana test

Michael TontchevMar 23, 2023, 7:12 AM
23 points
6 comments4 min readLW link

A method for em­piri­cal back-test­ing of AI’s abil­ity to self-improve

Michael TontchevMar 21, 2023, 8:24 PM
3 points
0 comments2 min readLW link

Paper­clipGPT(-4)

Michael TontchevMar 14, 2023, 10:03 PM
7 points
0 comments11 min readLW link