A third year computer engineering student studying at FRCRCE, Mumbai, India.
Ashutosh
Karma: 46
https://openai.com/index/disrupting-malicious-uses-of-ai-prc-linked-abuse/
I mean how do people forget things like this, openai looks at userdata all the time via automated or non automated means, it’s practically impossible for them to be so helpful on these edge cases given the traffic they get, including that school shooter example. Now they likely trained on buckmaster’s data, accessing thread is debatable.
gemini 3.5 flash’s reasoning summary model in antigravity must be of this type, it’s too optimistic!
I don’t have source for this on hand, but I think the word ‘e/acc’ is trying to mock ‘E-A’, it rhymes with e-yuck.
In my view it is anti-human at worst—or post-human. To quote Meditations On Moloch.
“Gnon” is Nick Land’s shorthand for “Nature And Nature’s God”, except the A is changed to an O and the whole thing is reversed, because Nick Land react to comprehensibility the same way as vampires to sunlight.
Land argues that humans should be more Gnon-conformist (pun Gnon-intentional). He says we do all these stupid things like divert useful resources to feed those who could never survive on their own, or supporting the poor in ways that encourage dysgenic reproduction, or allowing cultural degeneration to undermine the state. This means our society is denying natural law, basically listening to Nature say things like “this cause has this effect” and putting our fingers in our ears and saying “NO IT DOESN’T”. Civilizations that do this too much tend to decline and fall, which is Gnon’s fair and dispassionately-applied punishment for violating His laws.
[...]
More Lovecraft: the Internet popularization of the Cthulhu Cult claims that if you help free Cthulhu from his watery grave, he will reward you by eating you first, thus sparing you the horror of seeing everyone else eaten. This is a misrepresentation of the original text. In the original, his cultists receive no reward for freeing him from his watery prison, not even the reward of being killed in a slightly less painful manner.
The thought that abstract ideas can be Lovecraftian monsters is an old one but a deep one.
On the margin, compliance with the Gods of the Copybook Headings, Gnon, Cthulhu, whatever, may buy you slightly more time than the next guy. But then again, it might not. And in the long run, we’re all dead and our civilization has been destroyed by unspeakable alien monsters.
At some point, somebody has to say “You know, maybe freeing Cthulhu from his watery prison is a bad idea. Maybe we should not do that.”
That person will not be Nick Land. He is totally one hundred percent in favor of freeing Cthulhu from his watery prison and extremely annoyed that it is not happening fast enough. I have such mixed feelings about Nick Land. On the grail quest for the True Futurology, he has gone 99.9% of the path and then missed the very last turn, the one marked ORTHOGONALITY THESIS.
But the thing about grail quests is – if you make a wrong turn two blocks away from your house, you end up at the corner store feeling mildly embarrassed. If you do almost everything right and then miss the very last turn, you end up being eaten by the legendary Black Beast of Aaargh whose ichorous stomach acid erodes your very soul into gibbering fragments.
As far as I can tell from reading his blog, Nick Land is the guy in that terrifying border region where he is smart enough to figure out several important arcane principles about summoning demon gods, but not quite smart enough to figure out the most important such principle, which is NEVER DO THAT.
There’s also this lecture titled “e/acc has no theory” (sarcasm).
Another evidence, their unofficial discord server with 2-3k members is full of scam links.
I just stumbled upon that right now from their github, included it in the original message, I think it might fare similarly in official bench, it’s not mythos class yet.
The only benchmark I trust is deepswe. Better methodology, actually matches the feel in SWE tasks, tries to avoid benchmark poisoning, prevents models from cheating, allows them to write tests etc. GlM 5.1 is not that high up in there. They’re yet to test GLM 5.2, but I would guess it’s somewhere near 5.4 mini or optimistically between opus 4.6 high and 4.7 on max.
edit: I looked further into it GLM 5.2′s launch page . Has it listed at 46.2. Which puts it between 4.6 high and 4.7 max.
someone at Anthropic, or someone who has strong influence over Anthropic’s decisions
Not that it’s likely but, 80% of their code is written by mythos, their code 8x’ed since mythos, the model can sabotage this if it escapes review, bun rewrite was also done via it, I have been getting various segfaults on opencode ever since that happened—too lazy to report them to anthropic/bun team.
reciprocally selfish motivated by infinite drowning children and sense of duty towards oneself first would be 2?
Meditations On Moloch This is the dream time Forages vs farmers. Why the age of elm will happen were the posts which led me to similar conclusions as the author, needless to say AI causing painless extinction is much more bearable of a future than Moloch trap killing humans from inside. Maybe I am not hansonian enough to find these prospects comforting and I don’t think the eternal subsistence of “poor people smiling” equilibrium will last. I have lost weeks of sleep over this.
the idea of human life as the most important value.
It’s very difficult it’s sort of counter intuitive to people, they’re fine with saving lives in a vaccum, but when you have adversarial dynamics between cultures, going to war seems heroic and important. Moreover the political reality is not of lives of equal value, value of your life depends mostly on where and to whom you were born, and what power you wield.
This is what I tell people, you don’t get to have preservative less food of perfect preservation without it being economically feasible, you get to have stale food, tradeoffs to everything.
The way I experience ‘love’ the way I described above there is no clear love more or less of a particular trait.
Does that mean love has no impact on your decision making? Say you love two people equally for different traits, competence or kindess, take your example you have a trolley running towards Jack/John and you could switch it towards John/Jack, will you abstain from such a decision due to lack of preferential treatment or is love just a thing up in the air with no impact on your inclinations?
Ashutosh’s Shortform
The status related emotions are quite difficult to manage, one way I like to think of it is proportionality rather than equality, I know you probably don’t want to give up on SJW stuff, but textbook case for equality was getting too unbearable for me, I live in mumbai and it’s not possible for me to stay in a “bubble”, I only judge a person’s skill level proportional to who they’re and from where they have come, it would be marvelous for an ant to use lever about a fulcrum but for a modern human being it’s quite mundane. A person coming from bottom of the barrel to top is much more exceptional and worthy of attention, rather than an elite passing on the flag in their relay race.
There’re a lot of other status related emotions that I struggle with, of course the rationalist sin of underconfidence is one of them, it’s almost as if my attempt to not be elitists cuts in the opposite direction—ranking myself as low status, which makes socializing difficult. I am still figuring this out. I am no elite by any means, but being better at some things makes you realize how horrible others are at that thing , so much so that you can start questioning yourself on grounds of social conformity (modest epistemology) that maybe it’s the other way around.
The reason I liked multiplayer was because single-player felt too contrived, eerie and lonely. I didn’t felt alive I was the only actor in it who had immense agency.