RSS

Thomas Kwa

Karma: 3,237

AI safety researcher

Mislead­ing phrase in a GiveWell Youtube ad

Thomas KwaJan 5, 2023, 10:28 AM
86 points
13 comments1 min readEA link

EA fo­rum con­tent might be de­clin­ing in qual­ity. Here are some pos­si­ble mechanisms.

Thomas KwaSep 24, 2022, 10:24 PM
130 points
46 comments2 min readEA link

Most prob­lems fall within a 100x tractabil­ity range (un­der cer­tain as­sump­tions)

Thomas KwaMay 4, 2022, 12:06 AM
116 points
25 comments4 min readEA link

How dath ilan co­or­di­nates around solv­ing AI alignment

Thomas KwaApr 14, 2022, 1:53 AM
13 points
1 comment5 min readEA link

The case for in­fant outreach

Thomas KwaApr 1, 2022, 4:25 AM
246 points
8 comments4 min readEA link

Can we simu­late hu­man evolu­tion to cre­ate a some­what al­igned AGI?

Thomas KwaMar 29, 2022, 1:23 AM
19 points
0 comments7 min readEA link

Effec­tive­ness is a Con­junc­tion of Multipliers

Thomas KwaMar 25, 2022, 6:44 PM
259 points
34 comments4 min readEA link

Penn EA Res­i­dency Takeaways

Thomas KwaNov 12, 2021, 9:34 AM
99 points
10 comments9 min readEA link

“Hinge of His­tory” Refuted

Thomas KwaApr 1, 2021, 7:00 AM
174 points
6 comments6 min readEA link

Thomas Kwa’s Quick takes

Thomas KwaSep 23, 2020, 7:25 PM
2 points
82 comments1 min readEA link