RSS

alexflint

Karma: 2,238

Independent AI safety researcher

The ground of optimization

alexflint20 Jun 2020 0:38 UTC
159 points
68 comments27 min readLW link

Our take on CHAI’s re­search agenda in un­der 1500 words

alexflint17 Jun 2020 12:24 UTC
95 points
19 comments5 min readLW link

Search ver­sus design

alexflint16 Aug 2020 16:53 UTC
83 points
39 comments36 min readLW link

Reflec­tions on Larks’ 2020 AI al­ign­ment liter­a­ture review

alexflint1 Jan 2021 22:53 UTC
77 points
8 comments6 min readLW link

Re­ply to Paul Chris­ti­ano on Inac­cessible Information

alexflint5 Jun 2020 9:10 UTC
76 points
15 comments6 min readLW link

Pars­ing Chris Min­gard on Neu­ral Networks

alexflint6 May 2021 22:16 UTC
62 points
26 comments6 min readLW link

Agency in Con­way’s Game of Life

alexflint13 May 2021 1:07 UTC
61 points
66 comments9 min readLW link

My take on Michael Littman on “The HCI of HAI”

alexflint2 Apr 2021 19:51 UTC
57 points
4 comments7 min readLW link

Con­cern­ing not get­ting lost

alexflint14 May 2021 19:38 UTC
49 points
9 comments4 min readLW link

Un­der­stand­ing the Lot­tery Ticket Hy­poth­e­sis

alexflint14 May 2021 0:25 UTC
47 points
9 comments8 min readLW link

Life and ex­pand­ing steer­able consequences

alexflint7 May 2021 18:33 UTC
46 points
2 comments4 min readLW link

Where are in­ten­tions to be found?

alexflint21 Apr 2021 0:51 UTC
44 points
12 comments9 min readLW link

Prob­a­bil­ity the­ory and log­i­cal in­duc­tion as lenses

alexflint23 Apr 2021 2:41 UTC
37 points
7 comments6 min readLW link

Thoughts on Ia­son Gabriel’s Ar­tifi­cial In­tel­li­gence, Values, and Alignment

alexflint14 Jan 2021 12:58 UTC
35 points
14 comments4 min readLW link

Op­por­tu­ni­ties for in­di­vi­d­ual donors in AI safety

alexflint31 Mar 2018 18:37 UTC
30 points
3 comments11 min readLW link

Su­per­ra­tional­ity and net­work flow control

alexflint22 Jul 2013 1:49 UTC
29 points
11 comments1 min readLW link

Feed­back is cen­tral to agency

alexflint1 Jun 2020 12:56 UTC
28 points
0 comments3 min readLW link

Be­ware over-use of the agent model

alexflint25 Apr 2021 22:19 UTC
28 points
9 comments5 min readLW link

Knowl­edge is not just map/​ter­ri­tory resemblance

alexflint25 May 2021 17:58 UTC
28 points
4 comments3 min readLW link