AI Safety Newslet­ter #4: AI and Cy­ber­se­cu­rity, Per­sua­sive AIs, Weaponiza­tion, and Ge­offrey Hin­ton talks AI risks

Center for AI SafetyMay 2, 2023, 4:51 PM
35 points
2 comments5 min readEA link
(newsletter.safe.ai)

Si­mu­lat­ing a pos­si­ble al­ign­ment solu­tion in GPT2-medium us­ing Archety­pal Trans­fer Learning

MiguelMay 2, 2023, 4:23 PM
4 points
0 comments18 min readEA link

Re­view of The Good It Promises, the Harm It Does

Richard Y Chappell🔸May 2, 2023, 3:45 PM
217 points
55 comments10 min readEA link
(rychappell.substack.com)

Ap­ply Now: First-Ever EAGxNYC This August

Arthur Malone🔸May 2, 2023, 3:24 PM
101 points
4 comments2 min readEA link

Le­gal Pri­ori­ties Pro­ject – An­nual Re­port 2022

Legal Priorities ProjectMay 2, 2023, 1:32 PM
76 points
2 comments30 min readEA link

P(doom|AGI) is high: why the de­fault out­come of AGI is doom

Greg_Colbourn ⏸️ May 2, 2023, 10:40 AM
13 points
28 comments3 min readEA link

AGI ris­ing: why we are in a new era of acute risk and in­creas­ing pub­lic aware­ness, and what to do now

Greg_Colbourn ⏸️ May 2, 2023, 10:17 AM
68 points
35 comments13 min readEA link

An­nounc­ing Two Events around EAG Lon­don in Col­lab­o­ra­tion with the STEM (Science, Tech­nol­ogy, Eng­ineer­ing, and Math­e­mat­ics) Communities

Jessica WenMay 2, 2023, 9:14 AM
23 points
0 comments3 min readEA link

AGI safety ca­reer advice

richard_ngoMay 2, 2023, 7:36 AM
211 points
20 comments13 min readEA link

Sum­maries of top fo­rum posts (24th − 30th April 2023)

Zoe WilliamsMay 2, 2023, 2:30 AM
44 points
0 comments10 min readEA link

Owain Evans on LLMs, Truth­ful AI, AI Com­po­si­tion, and More

Ozzie GooenMay 2, 2023, 1:20 AM
21 points
0 comments1 min readEA link
(quri.substack.com)

Ex­plor­ing Me­tac­u­lus’s AI Track Record

Peter ScoblicMay 1, 2023, 9:02 PM
52 points
5 comments5 min readEA link

Call for Pythia-style foun­da­tion model suite for al­ign­ment research

LucretiaMay 1, 2023, 8:26 PM
10 points
0 comments1 min readEA link

The costs of caution

Kelsey PiperMay 1, 2023, 8:04 PM
112 points
17 comments4 min readEA link

[Linkpost] ‘The God­father of A.I.’ Leaves Google and Warns of Danger Ahead

imp4rtial 🔸May 1, 2023, 7:54 PM
43 points
3 comments3 min readEA link
(www.nytimes.com)

LessWrong Com­mu­nity Week­end 2023 [Ap­pli­ca­tions now closed]

LWCW2023May 1, 2023, 7:07 PM
3 points
0 comments6 min readEA link

LessWrong Com­mu­nity Week­end 2023 [Ap­pli­ca­tions now closed]

LWCW2023May 1, 2023, 7:07 PM
5 points
0 comments6 min readEA link

Ret­ro­spec­tive on re­cent ac­tivity of Ries­gos Catas­trófi­cos Globales

Jaime SevillaMay 1, 2023, 6:35 PM
45 points
0 comments5 min readEA link

List of AI safety newslet­ters and other resources

LizkaMay 1, 2023, 5:24 PM
49 points
5 comments4 min readEA link

New Nu­clear Se­cu­rity Syl­labus + Sum­mer Course

Maya DMay 1, 2023, 5:02 PM
45 points
5 comments1 min readEA link

My cur­rent take on ex­is­ten­tial AI risk [FB post]

Aryeh EnglanderMay 1, 2023, 4:22 PM
10 points
0 comments3 min readEA link

In fa­vor of steelmanning

JP Addison🔸May 1, 2023, 3:33 PM
27 points
3 comments3 min readEA link

Overview: Reflec­tion Pro­jects on Com­mu­nity Reform

Joris 🔸May 1, 2023, 3:14 PM
67 points
5 comments4 min readEA link

In­ter­me­di­ate goals for re­duc­ing risks from nu­clear weapons: A shal­low re­view (part 1/​4)

MichaelA🔸May 1, 2023, 3:04 PM
35 points
0 comments11 min readEA link
(docs.google.com)

Should EA grant­mak­ers make p(suc­cess) pub­lic?

KaleemMay 1, 2023, 1:42 PM
26 points
3 comments3 min readEA link

Char­ity Feed­back from 2022 Char­ity Evaluations

Animal Charity EvaluatorsMay 1, 2023, 10:57 AM
29 points
1 comment4 min readEA link
(animalcharityevaluators.org)

Global com­put­ing capacity

Vasco Grilo🔸May 1, 2023, 6:09 AM
12 points
0 comments1 min readEA link
(aiimpacts.org)

Could Ukraine re­take Crimea?

mhint199May 1, 2023, 1:06 AM
6 points
3 comments4 min readEA link

[Question] Values of a Space-Far­ing Civilization

Anthony FlemingApr 30, 2023, 10:17 PM
7 points
1 comment1 min readEA link

Dis­cus­sion about AI Safety fund­ing (FB tran­script)

AkashApr 30, 2023, 7:05 PM
104 points
10 comments6 min readEA link

First clean wa­ter, now clean air

finmApr 30, 2023, 6:01 PM
189 points
14 comments17 min readEA link
(finmoorhouse.com)

Bridg­ing EA’s Gen­der Gap: In­put From 60 Peo­ple

Alexandra BosApr 30, 2023, 4:20 PM
82 points
35 comments7 min readEA link

Con­nec­tomics seems great from an AI x-risk perspective

Steven ByrnesApr 30, 2023, 2:38 PM
10 points
0 comments10 min readEA link

Ca­reer un­cer­tainty: Medicine vs. AI

Markus KöthApr 30, 2023, 8:41 AM
20 points
9 comments1 min readEA link

Call for sub­mis­sions: Choice of Fu­tures sur­vey questions

c.troutApr 30, 2023, 6:59 AM
11 points
0 comments2 min readEA link
(airtable.com)

Sur­vival and Flour­ish­ing Fund’s 2023 H1 recs

AustinApr 30, 2023, 4:35 AM
39 points
2 comments2 min readEA link
(survivalandflourishing.fund)

In­tro­duc­ing Stan­ford’s new Hu­mane & Sus­tain­able Food Lab

MMathur🔸Apr 30, 2023, 1:14 AM
85 points
12 comments5 min readEA link

If you’d like to do some­thing about sex­ual mis­con­duct and don’t know what to do…

Habiba BanuApr 30, 2023, 1:09 AM
250 points
4 comments22 min readEA link

[Question] Are you op­ti­mistic about the com­mer­cial­iza­tion of alt pro­teins in 2023 and be­yond?

Eevee🔹Apr 29, 2023, 10:20 PM
24 points
7 comments1 min readEA link

Re­search agenda: Su­per­vis­ing AIs im­prov­ing AIs

Quintin PopeApr 29, 2023, 5:09 PM
16 points
0 comments19 min readEA link

Re­source: read­ings to learn more about HR/​peo­ple op­er­a­tions work

JosephApr 29, 2023, 2:05 PM
12 points
1 comment1 min readEA link

More global warm­ing might be good to miti­gate the food shocks caused by abrupt sun­light re­duc­tion scenarios

Vasco Grilo🔸Apr 29, 2023, 8:24 AM
46 points
39 comments13 min readEA link

A Guide to Fore­cast­ing AI Science Capabilities

Eleni_AApr 29, 2023, 6:51 AM
19 points
1 comment4 min readEA link

[SEE NEW EDITS] No, *You* Need to Write Clearer

Nicholas KrossApr 29, 2023, 5:04 AM
71 points
8 comments5 min readEA link
(www.thinkingmuchbetter.com)

Up­dated ‘Psy­chol­ogy of EA’ course: read­ing, videos, and syllabus

Geoffrey MillerApr 28, 2023, 8:43 PM
46 points
4 comments12 min readEA link

EA for Jews—Events and Op­por­tu­ni­ties

EA for JewsApr 28, 2023, 6:30 PM
24 points
0 comments1 min readEA link

[Question] Weigh­ing in solipsism

Liam 🔸Apr 28, 2023, 4:57 PM
3 points
3 comments1 min readEA link

Bet­ter weather fore­cast­ing: Agri­cul­tural and non-agri­cul­tural benefits in low- and lower-mid­dle-in­come countries

Rethink PrioritiesApr 28, 2023, 4:25 PM
41 points
6 comments3 min readEA link
(rethinkpriorities.org)

[Question] What are some (po­ten­tially) effec­tive policy ad­vo­cacy or­ga­ni­za­tions in global health and de­vel­op­ment?

Maxim VandaeleApr 28, 2023, 3:48 PM
24 points
5 comments1 min readEA link

New open let­ter on AI — “In­clude Con­scious­ness Re­search”

Jamie_HarrisApr 28, 2023, 7:50 AM
55 points
1 comment3 min readEA link
(amcs-community.org)