Elliott Thornley
@elliottthornley.bsky.social
đ€ 265
đ„ 407
đ 78
Research Fellow at Oxford University's Global Priorities Institute. Working on the philosophy of AI.
Recent article on the POST-Agents Proposal!
loading . . .
Shutdownable Agents through POST-Agency â LessWrong
Summary * Future artificial agents might resist shutdown. * I present an idea â the POST-Agents Proposal â for ensuring that doesnât happen. * I pâŠ
https://www.lesswrong.com/posts/JuRdvZyqaFbvTPemn/shutdownable-agents-through-post-agency-1
18 days ago
1
3
0
reposted by
Elliott Thornley
Global Priorities Institute
4 months ago
A new working paper, "Shutdownable Agents through POST-Agency" by Elliott Thornley, is now available on our website. Read it here:
globalprioritiesinstitute.org/thornley-shu...
loading . . .
Shutdownable Agents through POST-Agency - Elliott Thornley
Many fear that future artificial agents will resist shutdown. I present an idea â the POST-Agents Proposal â for ensuring that doesnât happen. I propose that we train agents to satisfy Preferences Onl...
https://globalprioritiesinstitute.org/thornley-shutdownable-agents-through-post-agency/
0
2
1
'Where are you?' seems like a pretty normal question, but for 99.99% of human history it basically never made sense to ask it.
5 months ago
0
5
0
Daniel Kerson wrote a nice summary of a talk I gave in Singapore last month
kerson.ai/how-advanced...
loading . . .
How Advanced AI Agents Could Resist Shutdown and What Can Be Done - Kerson AI Solutions
I attended a talk today at the SASH (Singapore AI Safety Hub). The speaker today was Elliott Thornley who is a Research Fellow at Oxford University. Introduction As artificial intelligence advances, w...
https://kerson.ai/how-advanced-ai-agents-could-resist-shutdown-and-what-can-be-done/
6 months ago
0
2
0
Our poster for TAIS 2025
6 months ago
0
3
0
A gif we made summarizing our 'Towards shutdownable agents' paper for TAIS 2025.
loading . . .
6 months ago
0
1
0
Gave a talk about the shutdown problem at the new Singapore AI Safety Hub!
6 months ago
2
6
0
I've got a new paper out open-access in AJP! Itâs about critical-level and critical-range views in population axiology, and why I think theyâre troubled by questions of identity between lives.
www.tandfonline.com/doi/full/10....
loading . . .
Critical-Set Views, Biographical Identity, and the Long Term
Critical-set views avoid the Repugnant Conclusion by subtracting some constant from the welfare score of each life in a population. These views are thus sensitive to facts about biographical identi...
https://www.tandfonline.com/doi/full/10.1080/00048402.2025.2476692
7 months ago
0
3
0
Progress in AI has been rapid in recent years. By contrast, progress in 'opening sentences of papers about AI' has completely stalled.
10 months ago
0
9
0
reposted by
Elliott Thornley
"Sure, the last 1000 grad students failed to solve the problem of induction, but that's no reason to think I can't do it."
10 months ago
3
34
14
reposted by
Elliott Thornley
Global Priorities Institute
10 months ago
Weâre excited to announce our new research agendas â for philosophy, economics and psychology â have now been published! You can read them here:
globalprioritiesinstitute.org/research-age...
loading . . .
Research agenda - Global Priorities Institute
The central focus of GPI is what we call âglobal priorities researchâ: research into issues that arise in response to the question, âWhat should we do with a given amount of limited resources if our a...
https://globalprioritiesinstitute.org/research-agenda/
0
19
7
[Pasting over an old Twitter thread about this post.]
loading . . .
The Shutdown Problem: Incomplete Preferences as a Solution â AI Alignment Forum
Preamble This post is an updated explanation of the Incomplete Preferences Proposal (IPP): my proposed solution to the shutdown problem. The post isâŠ
https://www.alignmentforum.org/posts/YbEbwYWkf8mv9jnmi/the-shutdown-problem-incomplete-preferences-as-a-solution
10 months ago
1
2
0
loading . . .
The introduction to my PhD thesis
[You can read it as a PDF here.]
https://openairopensea.substack.com/p/the-introduction-to-my-phd-thesis
11 months ago
0
3
0
Minor updates to an old post!
openairopensea.substack.com/p/my-favouri...
loading . . .
My favourite arguments against person-affecting views
1.
https://openairopensea.substack.com/p/my-favourite-arguments-against-person
11 months ago
0
3
0
Paper!
arxiv.org/pdf/2407.00805
With Alex Roman, Christos Ziakas, Leyton Ho, and Louis Thomson. Quick thread explaining it.
loading . . .
https://arxiv.org/pdf/2407.00805
11 months ago
1
2
1
11 months ago
0
0
0
Oh no
11 months ago
0
0
0
you reached the end!!
feeds!
log in
