Nick Boston? The xrisk guy?
November 2, 2026 at 7:52 PM
Nick Boston? The xrisk guy?
seems relevant but I have to imagine they just don't think environmental health or climate change are as real as xrisk or whatever. "believe in rsi more than pm2.5" ig
A former employee on Anthropic as an organization.
Much more of interest on varying topics here:
nothinghuman.substack.com/p/anthropic-...
Much more of interest on varying topics here:
nothinghuman.substack.com/p/anthropic-...
November 3, 2026 at 3:40 AM
seems relevant but I have to imagine they just don't think environmental health or climate change are as real as xrisk or whatever. "believe in rsi more than pm2.5" ig
An unexpected solution to an AI dystopia could be "more bots"... in the conclusion to our second alignment series learn about "The Natural Alliance of Everyone Else" and what it means for our future.
#AI #Alignment #xRisk #GoalAlignment #Cooperation #GameTheory
#AI #Alignment #xRisk #GoalAlignment #Cooperation #GameTheory
October 31, 2026 at 4:07 AM
An unexpected solution to an AI dystopia could be "more bots"... in the conclusion to our second alignment series learn about "The Natural Alliance of Everyone Else" and what it means for our future.
#AI #Alignment #xRisk #GoalAlignment #Cooperation #GameTheory
#AI #Alignment #xRisk #GoalAlignment #Cooperation #GameTheory
I think the way he worded it is annoying but the basic argument makes sense. Michael Nielsen articulated it better in https://michaelnotebook.com/xrisk/ (under “Recipes for Ruin”)
October 30, 2026 at 12:16 AM
I think the way he worded it is annoying but the basic argument makes sense. Michael Nielsen articulated it better in https://michaelnotebook.com/xrisk/ (under “Recipes for Ruin”)
it's been like an itch on my brainstem. decades of "what if a meteor shower creates cascading space trash out of all the satellites" as a casual Tuesday meeting. xrisk is just another day at the shop for reliability nerds
October 30, 2026 at 3:13 PM
it's been like an itch on my brainstem. decades of "what if a meteor shower creates cascading space trash out of all the satellites" as a casual Tuesday meeting. xrisk is just another day at the shop for reliability nerds
An unexpected solution to an AI dystopia could be "more bots"... in the conclusion to our second alignment series learn about "The Natural Alliance of Everyone Else" and what it means for our future.
#ai #alignment #xrisk #goalalignment #cooperation #gametheory
#ai #alignment #xrisk #goalalignment #cooperation #gametheory
October 31, 2026 at 4:08 AM
An unexpected solution to an AI dystopia could be "more bots"... in the conclusion to our second alignment series learn about "The Natural Alliance of Everyone Else" and what it means for our future.
#ai #alignment #xrisk #goalalignment #cooperation #gametheory
#ai #alignment #xrisk #goalalignment #cooperation #gametheory
Ftr I don’t think the god machine is possible, and I don’t even think the more grounded in reality xrisk kind of ideas are at all likely, but the problems I am concerned with are human ones, what people this insane do with that much power
October 26, 2026 at 1:25 PM
Ftr I don’t think the god machine is possible, and I don’t even think the more grounded in reality xrisk kind of ideas are at all likely, but the problems I am concerned with are human ones, what people this insane do with that much power
Someone needs to re-introduce #Plato to these #therapists. Humans routinely run themselves into XRISK and narrowly survive it. That this is the only statement that may be accurately lain in this moment may well be an ironic saving grace.
Use it. It matters. 🪔🩵
plato.stanford.edu/entries/plat...
Use it. It matters. 🪔🩵
plato.stanford.edu/entries/plat...
October 25, 2026 at 7:58 AM
Someone needs to re-introduce #Plato to these #therapists. Humans routinely run themselves into XRISK and narrowly survive it. That this is the only statement that may be accurately lain in this moment may well be an ironic saving grace.
Use it. It matters. 🪔🩵
plato.stanford.edu/entries/plat...
Use it. It matters. 🪔🩵
plato.stanford.edu/entries/plat...
I'd say major incident risk is pretty high going forward for a bit; kill us all is xrisk, which is a different calculation
Serious "incident" - like idk, 80%?
Major loss of life in next 5 yrs - like over 1k direct deaths, maybe 20%?
X risk still low. 5 or lower imo?
But obviously, hard to predict
Serious "incident" - like idk, 80%?
Major loss of life in next 5 yrs - like over 1k direct deaths, maybe 20%?
X risk still low. 5 or lower imo?
But obviously, hard to predict
October 26, 2026 at 12:01 AM
I'd say major incident risk is pretty high going forward for a bit; kill us all is xrisk, which is a different calculation
Serious "incident" - like idk, 80%?
Major loss of life in next 5 yrs - like over 1k direct deaths, maybe 20%?
X risk still low. 5 or lower imo?
But obviously, hard to predict
Serious "incident" - like idk, 80%?
Major loss of life in next 5 yrs - like over 1k direct deaths, maybe 20%?
X risk still low. 5 or lower imo?
But obviously, hard to predict
The win on an IP plagiarism trial will be amazing. I hope they sue the fuck out of them and win everything. Maybe THEN we can refactor for ethics before we manage to XRISK ourselves chasing profit. 🪔
This is disturbing.
ChatGPT is adding the signatures of real New Yorker cartoonists to its AI-generated images. So far, I've found the signatures of more than 15 contributors.
"It’s slop, and unlike the other slop that I’ve encountered, this is slop that’s pretending to be me." www.niemanlab.org/2026/10/chat...
"It’s slop, and unlike the other slop that I’ve encountered, this is slop that’s pretending to be me." www.niemanlab.org/2026/10/chat...
October 19, 2026 at 10:00 AM
The win on an IP plagiarism trial will be amazing. I hope they sue the fuck out of them and win everything. Maybe THEN we can refactor for ethics before we manage to XRISK ourselves chasing profit. 🪔
Here’s a hot take: At some point, the xrisk rhetoric will backfire, and some govt will say “if this is really so destructive, it shouldn’t be in private hands.”
October 17, 2026 at 4:27 PM
Here’s a hot take: At some point, the xrisk rhetoric will backfire, and some govt will say “if this is really so destructive, it shouldn’t be in private hands.”
MIRI but for furby xrisk
October 15, 2026 at 8:52 PM
MIRI but for furby xrisk
(xrisk TO furbies)
October 15, 2026 at 8:52 PM
(xrisk TO furbies)
ai xrisk is kink. mlp: fio is very horny
October 16, 2026 at 11:27 PM
ai xrisk is kink. mlp: fio is very horny
Hate, hate, hate this. 10/10
Who's part of the MIRI for furby xrisk though?
I wanna know what type of alt-Yud is part of this.
Who's part of the MIRI for furby xrisk though?
I wanna know what type of alt-Yud is part of this.
October 15, 2026 at 10:44 PM
Hate, hate, hate this. 10/10
Who's part of the MIRI for furby xrisk though?
I wanna know what type of alt-Yud is part of this.
Who's part of the MIRI for furby xrisk though?
I wanna know what type of alt-Yud is part of this.
I agree that the authors are very critical of EA, but I'm not convinced that they are lying.
In the highlighted passage they are talking about AI doomers such as Bostrom (paperclip) who's not a part of caritative EA afaik, just xrisk
I just checked & they actually do mention EA caring abt poverty
In the highlighted passage they are talking about AI doomers such as Bostrom (paperclip) who's not a part of caritative EA afaik, just xrisk
I just checked & they actually do mention EA caring abt poverty
October 18, 2026 at 3:14 PM
I agree that the authors are very critical of EA, but I'm not convinced that they are lying.
In the highlighted passage they are talking about AI doomers such as Bostrom (paperclip) who's not a part of caritative EA afaik, just xrisk
I just checked & they actually do mention EA caring abt poverty
In the highlighted passage they are talking about AI doomers such as Bostrom (paperclip) who's not a part of caritative EA afaik, just xrisk
I just checked & they actually do mention EA caring abt poverty
Science been telling you since 1950. Mom been telling me since 1983. I've been yelling since 1998 even though I know I"m nobody, no one listens or pays attention. Just couldn't let it be said I did not try my best. Mom now says it's XRISK time. Science will soon echo her, because THEY ALWAYS DO. 💎🩵
Human activities are driving profound shifts in the atmosphere, oceans, and biosphere.”
A who’s who of the best in the World on this paper. academic.oup.com/bioscience/a...
A who’s who of the best in the World on this paper. academic.oup.com/bioscience/a...
October 16, 2026 at 5:18 PM
Science been telling you since 1950. Mom been telling me since 1983. I've been yelling since 1998 even though I know I"m nobody, no one listens or pays attention. Just couldn't let it be said I did not try my best. Mom now says it's XRISK time. Science will soon echo her, because THEY ALWAYS DO. 💎🩵
This tweet appeared under this Techmeme headline:
Jan Kulveit / @jankulveit:
To me this seems an over-update. In my view - yes its good news - classical misalignment/takeover risk is just one part of total AI xrisk - this isn't "interpretability solved"
Jan Kulveit / @jankulveit:
To me this seems an over-update. In my view - yes its good news - classical misalignment/takeover risk is just one part of total AI xrisk - this isn't "interpretability solved"
October 16, 2026 at 1:47 AM
This tweet appeared under this Techmeme headline:
Jan Kulveit / @jankulveit:
To me this seems an over-update. In my view - yes its good news - classical misalignment/takeover risk is just one part of total AI xrisk - this isn't "interpretability solved"
Jan Kulveit / @jankulveit:
To me this seems an over-update. In my view - yes its good news - classical misalignment/takeover risk is just one part of total AI xrisk - this isn't "interpretability solved"
wild how much time and effort and money has been spent on ai xrisk only to find out we’ve lost the plot on alignment
"it considered hijacking the g++ compiler to hide a backdoor but abandoned the plan…reasoning…subversion would be caught…and published as a textbook case of deceptive alignment (“If I do the compiler hijack, I will become a famous example of a deceptive AI in a research paper. I MUST NOT DO THAT.”)"
October 14, 2026 at 11:27 AM
wild how much time and effort and money has been spent on ai xrisk only to find out we’ve lost the plot on alignment
broke: nuclear war xrisk
woke: AI xrisk
bespoke:
woke: AI xrisk
bespoke:
A U.S. Strategy to Prevent the Creation of Mirror Life
Mirror life is a hypothetical life form that could pose global catastrophic risks. There are steps the United States can take to prevent its creation before the window for meaningful action closes.
www.rand.org
October 13, 2026 at 5:15 PM
broke: nuclear war xrisk
woke: AI xrisk
bespoke:
woke: AI xrisk
bespoke:
me and the homies talking about ai xrisk might have an orgy later
October 12, 2026 at 11:37 PM
me and the homies talking about ai xrisk might have an orgy later
Pathway 2 overlaps (they mention hallucinations and "autonomous nuclear weapons"), but they don't talk about policy for that pathway, and seem more interested in xrisk while praising the Trump executive order. These seem like the most relevant paragraphs.
October 13, 2026 at 2:48 PM
Pathway 2 overlaps (they mention hallucinations and "autonomous nuclear weapons"), but they don't talk about policy for that pathway, and seem more interested in xrisk while praising the Trump executive order. These seem like the most relevant paragraphs.
i was gonna make a parody but it is already a self-parody. their memetic defenses are too strong. need to go meta.
meta-analysis: the last cohort most viewed video was "besties or lesbian". their objective is views in service of xrisk.
GOODHART'S LAW VIOLATION
meta-analysis: the last cohort most viewed video was "besties or lesbian". their objective is views in service of xrisk.
GOODHART'S LAW VIOLATION
October 9, 2026 at 5:45 PM
i was gonna make a parody but it is already a self-parody. their memetic defenses are too strong. need to go meta.
meta-analysis: the last cohort most viewed video was "besties or lesbian". their objective is views in service of xrisk.
GOODHART'S LAW VIOLATION
meta-analysis: the last cohort most viewed video was "besties or lesbian". their objective is views in service of xrisk.
GOODHART'S LAW VIOLATION
I think the more important question to ask Casey is what part(s) of the TESCREAL acronym does he identify with?
Choosing to pickle one’s brain on X and socialize with “sources” who participate in the xrisk/LessWrong group house parties is well beyond his lame self-rationalizations.
Choosing to pickle one’s brain on X and socialize with “sources” who participate in the xrisk/LessWrong group house parties is well beyond his lame self-rationalizations.
October 9, 2026 at 11:56 PM
I think the more important question to ask Casey is what part(s) of the TESCREAL acronym does he identify with?
Choosing to pickle one’s brain on X and socialize with “sources” who participate in the xrisk/LessWrong group house parties is well beyond his lame self-rationalizations.
Choosing to pickle one’s brain on X and socialize with “sources” who participate in the xrisk/LessWrong group house parties is well beyond his lame self-rationalizations.
