RSS

abramdemski(Abram Demski)

Karma: 17,982

[Question] Why is o1 so de­cep­tive?

abramdemski27 Sep 2024 17:27 UTC
176 points
23 comments3 min readLW link

For­mal­iz­ing the In­for­mal (event in­vite)

abramdemski10 Sep 2024 19:22 UTC
42 points
0 comments1 min readLW link

In Defense of Open-Minded UDT

abramdemski12 Aug 2024 18:27 UTC
72 points
27 comments11 min readLW link

Leav­ing MIRI, Seek­ing Funding

abramdemski8 Aug 2024 18:32 UTC
267 points
19 comments2 min readLW link

Cir­cu­lar Reasoning

abramdemski5 Aug 2024 18:10 UTC
91 points
36 comments8 min readLW link

LLMs for Align­ment Re­search: a safety pri­or­ity?

abramdemski4 Apr 2024 20:03 UTC
144 points
24 comments11 min readLW link

Modern Trans­form­ers are AGI, and Hu­man-Level

abramdemski26 Mar 2024 17:46 UTC
219 points
88 comments5 min readLW link

Tech­nolo­gies and Ter­minol­ogy: AI isn’t Soft­ware, it’s… Deep­ware?

13 Feb 2024 13:37 UTC
40 points
10 comments8 min readLW link

Mean­ing & Agency

abramdemski19 Dec 2023 22:27 UTC
91 points
17 comments14 min readLW link

FixDT

abramdemski30 Nov 2023 21:57 UTC
56 points
14 comments14 min readLW link

Agent Boundaries Aren’t Markov Blan­kets. [Un­less they’re non-causal; see com­ments.]

abramdemski20 Nov 2023 18:23 UTC
82 points
11 comments2 min readLW link

Trans­la­tions Should Invert

abramdemski5 Oct 2023 17:44 UTC
48 points
19 comments3 min readLW link

[Question] Where might I di­rect promis­ing-to-me re­searchers to ap­ply for al­ign­ment jobs/​grants?

abramdemski18 Sep 2023 16:20 UTC
45 points
10 comments1 min readLW link

One Minute Every Moment

abramdemski1 Sep 2023 20:23 UTC
125 points
23 comments3 min readLW link

Prob­a­bil­is­tic Payor Lemma?

abramdemski19 Mar 2023 17:57 UTC
69 points
7 comments4 min readLW link

Teleose­man­tics!

abramdemski23 Feb 2023 23:26 UTC
81 points
26 comments6 min readLW link

Some Thoughts on AI Art

abramdemski25 Jan 2023 14:18 UTC
74 points
20 comments7 min readLW link

Con­tra Com­mon Knowledge

abramdemski4 Jan 2023 22:50 UTC
52 points
31 comments16 min readLW link

Talk­ing to God

abramdemski3 Jan 2023 20:14 UTC
30 points
7 comments2 min readLW link

Knottiness

abramdemski2 Jan 2023 22:13 UTC
43 points
4 comments2 min readLW link