RSS

Oliver Sourbut

Karma: 2,186

oliversourbut.net

I’m particularly interested in sustainable collaboration and the long-term future of value. I’d love to contribute to a safer and more prosperous future with AI! Always interested in discussions about axiology, x-risks, s-risks.

I enjoy meeting new perspectives and growing my understanding of the world and the people in it. I also love to read—let me know your suggestions! In no particular order, here are some I’ve enjoyed recently

Cooperative gaming is a relatively recent but fruitful interest for me. Here are some of my favourites

People who’ve got to know me only recently are sometimes surprised to learn that I’m a pretty handy trumpeter and hornist.

Ci­ta­tions Needed: Magic En­cy­clo­pe­dias to Save the World

Oliver Sourbut12 Jun 2026 15:35 UTC
40 points
3 comments5 min readLW link
(www.oliversourbut.net)

Lab Leaks, Black Holes, and Eggs: Epistemic Case Study Competition

4 Jun 2026 16:26 UTC
44 points
6 comments8 min readLW link
(flf.org)

The main im­pact from au­to­mated AI pro­duc­tion: con­cen­tra­tion of power?

Oliver Sourbut31 May 2026 20:42 UTC
20 points
2 comments7 min readLW link
(www.oliversourbut.net)

A (Slightly) Mechanis­tic The­ory for Ex­po­nen­tially In­creas­ing AI Time Hori­zons?

Oliver Sourbut24 May 2026 15:52 UTC
50 points
17 comments8 min readLW link
(www.oliversourbut.net)

Eng­ineer­ing a Safer World: Risk Model­ling — and Safety Eng­ineer­ing? — for AI Loss of Control

Oliver Sourbut17 May 2026 16:02 UTC
10 points
1 comment9 min readLW link
(www.oliversourbut.net)

Re­in­force­ment learn­ing scal­ing might in­cen­tivise hid­den rea­son­ing ar­chi­tec­tures for AI

Oliver Sourbut10 May 2026 15:30 UTC
19 points
5 comments6 min readLW link
(www.oliversourbut.net)

How did ‘large’ lan­guage mod­els get that way? The role of Trans­form­ers and Pre­train­ing in GPT

Oliver Sourbut3 May 2026 21:35 UTC
16 points
0 comments7 min readLW link
(www.oliversourbut.net)

Is the Cat Out of the Bag?: Who knows how to make AGI?

Oliver Sourbut24 Apr 2026 21:49 UTC
35 points
0 comments4 min readLW link
(www.oliversourbut.net)

“Best hu­mans still out­perform”: One turn­ing point in the his­tory of cope around ar­tifi­cial intelligence

Oliver Sourbut17 Apr 2026 14:10 UTC
28 points
6 comments3 min readLW link
(www.oliversourbut.net)

Defense-favoured co­or­di­na­tion de­sign sketches

6 Apr 2026 15:19 UTC
18 points
6 comments25 min readLW link
(www.forethought.org)

Orders of mag­ni­tude: use semi­tones, not decibels

Oliver Sourbut1 Apr 2026 22:41 UTC
57 points
4 comments4 min readLW link
(www.oliversourbut.net)

Strate­gic aware­ness tools: de­sign sketches

11 Feb 2026 12:28 UTC
18 points
2 comments1 min readLW link
(www.forethought.org)

De­sign sketches for a more sen­si­ble world

9 Feb 2026 10:22 UTC
26 points
2 comments4 min readLW link
(www.forethought.org)

De­sign sketches for an­gels-on-the-shoulder

9 Feb 2026 9:52 UTC
23 points
0 comments2 min readLW link
(www.forethought.org)

AI for Hu­man Rea­son­ing for Rationalists

Oliver Sourbut3 Feb 2026 13:22 UTC
29 points
0 comments4 min readLW link
(www.oliversourbut.net)

[Question] Another Cost Disease? We are all cap­i­tal­ists now

Oliver Sourbut9 Jan 2026 13:07 UTC
16 points
11 comments2 min readLW link

A Full Epistemic Stack: Knowl­edge Com­mons for the 21st Century

19 Dec 2025 22:48 UTC
46 points
7 comments11 min readLW link
(www.oliversourbut.net)

Bet­ter than log­a­r­ith­mic re­turns to rea­son­ing?

Oliver Sourbut30 Jul 2025 0:50 UTC
14 points
5 comments3 min readLW link
(www.oliversourbut.net)

Do LLMs know what they’re ca­pa­ble of? Why this mat­ters for AI safety, and ini­tial findings

13 Jul 2025 19:54 UTC
53 points
5 comments18 min readLW link

You Can’t Skip Ex­plo­ra­tion: Why un­der­stand­ing ex­per­i­men­ta­tion and taste is key to un­der­stand­ing AI

Oliver Sourbut21 May 2025 16:08 UTC
20 points
0 comments11 min readLW link
(www.oliversourbut.net)