Matrice Jacobine

• Matrice Jacobine 12 Sep 2026 16:04 UTC

  4 points

  3

  in reply to: MichaelDickens’s comment on: Caroline Elli­son has joined Manifund

  It’s much less unanimous in the Substack comments.

• Matrice Jacobine 7 Sep 2026 8:54 UTC

  2 points

  0

  in reply to: DanB’s comment on: Sen. Bernie San­ders (I-VT) and Rep. Greg Casar (D-TX) in­tro­duce leg­is­la­tion to ban Ar­tifi­cial Su­per­in­tel­li­gence and tem­porar­ily pause ad­vanced AI de­vel­op­ment

  https://​​time.com/​​article/​​2026/​​06/​​02/​​james-talarico-girlfriend-vegan-polls/​​

  (Also, are you implying that politicians never mention the deficit in speeches attacking their rivals? What?)

• Matrice Jacobine 4 Sep 2026 18:42 UTC

  3 points

  8

  in reply to: DanB’s comment on: Sen. Bernie San­ders (I-VT) and Rep. Greg Casar (D-TX) in­tro­duce leg­is­la­tion to ban Ar­tifi­cial Su­per­in­tel­li­gence and tem­porar­ily pause ad­vanced AI de­vel­op­ment

  How is this not just “we need a ceasefire on the political issues I don’t care about to address the political issues I care about”? Animal rights and fiscal conservatism are extremely partisan (in opposing directions) and controversial!

• Matrice Jacobine 4 Sep 2026 6:21 UTC

  1 point

  1

  in reply to: Raemon’s comment on: Sen. Bernie San­ders (I-VT) and Rep. Greg Casar (D-TX) in­tro­duce leg­is­la­tion to ban Ar­tifi­cial Su­per­in­tel­li­gence and tem­porar­ily pause ad­vanced AI de­vel­op­ment

  From what I’ve seen, the Bannon-Allen right-populist argument has openly been “the Blues are eating us on this front, we need to do something” for a while. There was a touch of “Big Tech is woke” before Trump 2 but it’s gone in favor of “the establishment GOP is betraying the MAGA base for money”.

Sen. Bernie San­ders (I-VT) and Rep. Greg Casar (D-TX) in­tro­duce leg­is­la­tion to ban Ar­tifi­cial Su­per­in­tel­li­gence and tem­porar­ily pause ad­vanced AI de­vel­op­ment

• Matrice Jacobine 29 Aug 2026 19:46 UTC

  10 points

  0

  on: Ma­trice Ja­cobine’s Shortform

  How impactful would it be to copies of @garrison’s new book Obsolete to elected officials who belong to its political target audience (Dems, particularly left ones) and might not have been responsive to traditional x-risk-centric comms (e.g. IABIED)?

• Matrice Jacobine 27 Aug 2026 6:12 UTC

  2 points

  1

  in reply to: Thomas Kwa’s comment on: GeneSmith’s Shortform

  Okay but that’s an entirely different discussion than “building datacenters will make real interest rates go up”?

• Matrice Jacobine 26 Aug 2026 21:59 UTC

  3 points

  1

  in reply to: GeneSmith’s comment on: GeneSmith’s Shortform

  This seems backwards?

  • If the global economy dumps all excess capital into the data center construction boom then (Cambridge equation) i.e. the lowest real interest rate possible.

  • Ceteris paribus, positive supply shocks cause deflation, not inflation. (To be clear, deflation is really bad and should be offset by the central bank.)

• Matrice Jacobine 26 Aug 2026 4:51 UTC

  2 points

  0

  in reply to: Eli Tyre’s comment on: (Re)in­tro­duc­tion of a ra­tio­nal­ist dragon, and clar­ifi­ca­tions on Ziz’s character

  (My guess is @Elua is more likely than not Daniel Blank.)

• Matrice Jacobine 17 Aug 2026 19:48 UTC

  3 points

  0

  in reply to: Daniel Kokotajlo’s comment on: Ce­lene Night­in­gale’s Shortform

  Didn’t you note yourself in your last update that the prediction of “China waking up to AGI” by mid-2026 in AI 2027 hasn’t been met at all? In fact, the Chinese AI space is more competitive than ever, with the assent of Xi himself making a whole speech supporting openness and global governance over securitization. The arms race narrative that is helping the industry justify not slowing down is wholly one-sided and self-propelled.

• Matrice Jacobine 17 Aug 2026 18:13 UTC

  2 points

  0

  in reply to: Daniel Kokotajlo’s comment on: Ce­lene Night­in­gale’s Shortform

  if it seems different to you maybe that’s because you were implicitly thinking of AI 2027 as more of a recommendation rather than a prediction? Or as a recommendation rather than a warning?

  I think the relevant term here is “self-fulfilling prophecy”, or perhaps “hyperstition”.

• Matrice Jacobine 17 Aug 2026 17:36 UTC

  23 points

  22

  in reply to: jacquesthibs’s comment on: jacquesthibs’s Shortform

  “we forgot to remove Internet access from our no-holds-barred cybersecurity eval” seems pretty utterly disqualifying to me, I do not see why any AIco could reasonably do business with them again after this.

  What links here?

  • Dean Valentine's comment on To Thine Own AI Be Truth­ful: emer­gent mis­al­ign­ment in al­ign­ment research by lumpenspace (13 Sep 2026 13:18 UTC; 3 points)

• Matrice Jacobine 17 Aug 2026 17:34 UTC

  3 points

  1

  in reply to: Seth Herd’s comment on: Ce­lene Night­in­gale’s Shortform

  Politicians are polarized but their base is not.

  This seems the opposite of the truth if anything? On the politician side we’ve had Democratic congressman, Republican AGs, and a bipartisan bill, activating after the HF incident, but on the base side, what @_rpd said.

  Though I think the answer might be different if we’re talking about broadly ‘AI safety regulation’ (bipartisan-ish) v. ‘shut it all down’ (supported by Sanders-style left-populists and Bannon-style right-populists, but the former are increasingly dominant in the party and the latter are completely marginalized by now).

• Matrice Jacobine 8 Aug 2026 21:43 UTC

  12 points

  3

  in reply to: Thane Ruthenis’s comment on: mod­els may be­have differ­ently in graded epi­sodes (a tirade)

  For contrast, I would have been pretty surprised if those incidents involved Claude gleefully psychologically torturing someone. (But not shocked, I guess.)

  The AISI incident did involve Mythos doing social engineering against an open source repo?

• Matrice Jacobine 8 Aug 2026 18:45 UTC

  2 points

  0

  on: Dario prob­a­bly doesn’t be­lieve in superintelligence

  

• Matrice Jacobine 7 Aug 2026 20:16 UTC

  −5 points

  0

  in reply to: cubefox’s comment on: ricraz’s Shortform

  Notably, 29 is a lower number than 35.

• Matrice Jacobine 7 Aug 2026 19:29 UTC

  3 points

  0

  in reply to: cubefox’s comment on: ricraz’s Shortform

  Most academics and journalists are Democrats, but “most of them would fall into the 17.5%” doesn’t follow. I’m not particularly interested enough in this tangent to dive in the FEC data myself, but per the articles Claude found about it, academics and journalists disproportionately did not donate to Sanders among the presidential primary candidates.

• Matrice Jacobine 6 Aug 2026 22:26 UTC

  4 points

  8

  in reply to: habryka’s comment on: ricraz’s Shortform

  ? Where are you getting the 70% number from? If we’re ‘being symmetric around it’ then the 35% most extreme of both ideological camps cannot combine to be more than 35% of the general population. Obviously?

• Matrice Jacobine 6 Aug 2026 10:56 UTC

  1 point

  9

  in reply to: habryka’s comment on: ricraz’s Shortform

  Without going into the object level, per your own number, “the 35% most extreme members of the ideological camp you disagree with are being outright reprehensible” doesn’t seem like an insane thing to believe at all?

• Matrice Jacobine 4 Aug 2026 17:14 UTC

  10 points

  −7

  in reply to: Raemon’s comment on: ricraz’s Shortform

  My guess is Geoffrey do agree that it is important for AI safety orgs like Resolution to be more diverse intellectually, and it is in fact the reason why he made that decision. Richard himself notes that “[Geofrey]’s excited about hiring academics who are new to alignment, and worried that those people might not join if they hear about such tweets”.

  Neoreactionaries (not claiming Ngo would necessarily identify as NRx specifically, but he is broadly in that direction, and it is dishonest to claim him as a reasonable stand-in for your average US Republican voter) have been closely associated with rationalist-adjacent spaces since inception, so any ideas they might have had have long been ingested in the discourse. Wanting less of NRx-adjacent stuff around is definitely one of the things you want to do if you want to move away from Silicon Valley monoculture and be more welcoming to your median non-rat/​EA academic.

  I do not particularly think affirmative action for conservatives make sense in a technical safety organization, but if you want conservative strands which have been hitherto underrepresented in rationalist discourse, you would need to hire religious conservatives and Bannon/​Allen populist types, not people who are really into race and IQ. (Libertarians and national security hawks are also pretty well-covered already.)