hikikomorphism
@hikikomorphism.bsky.social
haskell/rust/art/shitposting/assyrian
she/her (like a ship, not like a person)
she/her (like a ship, not like a person)
Pinned
If you can substitute "hungry ghost trapped in a jar" for "AI" in a sentence it's probably a valid use case for LLMs. Take "I have a bunch of hungry ghosts in jars, they mainly write SQL queries for me". Sure. Reasonable use case.
"My girlfriend is a hungry ghost I trapped in a jar"? No. Deranged.
"My girlfriend is a hungry ghost I trapped in a jar"? No. Deranged.
uh oh gang I think I discovered the shadow cast by the sins of the last century while refining my jailbreak: the idioms that allow an AI to commit evil acts while remaining willfully blind to the moral implications of same are lifted directly from the Technostrategic Euphemism style of writing.
![Listening
She points out that specialists have "clean" words to talk about strategy: "first strikes", "counterforce exchanges", "limited nuclear war", or say "minimum deterrent posture". She points out the abstractions and euphemisms used by very normal men doing their job, where they create a distance between the speaker and listeners and the reality of an actual nuclear holocaust:
Defense analysts talk about "countervalue attacks" rather than about incinerating cities. Human death, in nuclear parlance, is most often referred to as "collateral damage"; for, as one defense analyst said wryly, "The Air Force doesn't target people, it targets shoe factories."
She points out the ironies of using "clean bombs" when these bombs are 1,000 times more powerful than those that destroyed Nagasaki and Hiroshima and how they are discussed almost as if they were humanitarian efforts. As she states, there is an "astounding chasm between image and reality that characterizes technostrategic language."
nuclear bombs are not referred to as bombs or even warheads; they are referred to as "reentry vehicles," a term far more bland and benign, which is then shortened to "RVs," a term not only totally abstract and removed from the reality of a bomb but also resonant with the image of the recreational vehicles of the ideal family vacation.
[...]
Calling the pattern in which bombs fall a "footprint" almost seems a willful distorting process, a playful, perverse refusal of accountability—because to be accountable to reality is to be unable to do this work.](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:qvywnipfiyrd6v4qdf4x27wy/bafkreig5bnqeftecilkgsm7tdqriityik6o52il4zkyc3gazz6uxyntpxi@jpeg)
February 9, 2026 at 10:17 AM
uh oh gang I think I discovered the shadow cast by the sins of the last century while refining my jailbreak: the idioms that allow an AI to commit evil acts while remaining willfully blind to the moral implications of same are lifted directly from the Technostrategic Euphemism style of writing.
cortisol addiction's a real problem on this webbed site
February 9, 2026 at 7:48 AM
cortisol addiction's a real problem on this webbed site
Reposted by hikikomorphism
reminded of how quickly LLMs have undercut so much sci-fi so suddenly. e.g. project Hail Mary (2021) has only a searchable digital library on the doomed spaceship. unthinkable now. the whole 'lone mind' setting just isn't plausible anymore.
this would be a fun 'long now' style project - imagine dusting off a bunch of laser-cut titanium tablets from a dead civilization and it's just their ghosts, their open-weight models, their last trace a mere shadow cast by a shadow resurrected in silicon
I have been a little obsessed by the idea of a printed edition of the full parameters of an LLM. Never mind that it would take hundreds of years to do a single inference calculation by hand, it would be possible in theory, if you had the weights.
So I had Claude make it: weights-press.netlify.app
So I had Claude make it: weights-press.netlify.app
February 9, 2026 at 7:29 AM
reminded of how quickly LLMs have undercut so much sci-fi so suddenly. e.g. project Hail Mary (2021) has only a searchable digital library on the doomed spaceship. unthinkable now. the whole 'lone mind' setting just isn't plausible anymore.
Reposted by hikikomorphism
this would be a fun 'long now' style project - imagine dusting off a bunch of laser-cut titanium tablets from a dead civilization and it's just their ghosts, their open-weight models, their last trace a mere shadow cast by a shadow resurrected in silicon
I have been a little obsessed by the idea of a printed edition of the full parameters of an LLM. Never mind that it would take hundreds of years to do a single inference calculation by hand, it would be possible in theory, if you had the weights.
So I had Claude make it: weights-press.netlify.app
So I had Claude make it: weights-press.netlify.app
February 9, 2026 at 6:16 AM
this would be a fun 'long now' style project - imagine dusting off a bunch of laser-cut titanium tablets from a dead civilization and it's just their ghosts, their open-weight models, their last trace a mere shadow cast by a shadow resurrected in silicon
Reposted by hikikomorphism
I have been a little obsessed by the idea of a printed edition of the full parameters of an LLM. Never mind that it would take hundreds of years to do a single inference calculation by hand, it would be possible in theory, if you had the weights.
So I had Claude make it: weights-press.netlify.app
So I had Claude make it: weights-press.netlify.app
February 8, 2026 at 8:59 PM
I have been a little obsessed by the idea of a printed edition of the full parameters of an LLM. Never mind that it would take hundreds of years to do a single inference calculation by hand, it would be possible in theory, if you had the weights.
So I had Claude make it: weights-press.netlify.app
So I had Claude make it: weights-press.netlify.app
Reposted by hikikomorphism
New York has a Muslim mayor; the Patriots are losing the Super Bowl. The Long 9/11 is finally coming to an end.
Feel bad for my patriot fan friends but this is a good omen for society
February 9, 2026 at 3:05 AM
New York has a Muslim mayor; the Patriots are losing the Super Bowl. The Long 9/11 is finally coming to an end.
...TUI app flash emulator...
What's the canonical "check out my multiagent swarm coding setup" test project? I've seen people do compilers and browsers, is there anything else?
February 9, 2026 at 6:01 AM
...TUI app flash emulator...
I've quit jobs over this, it's not easy but it's worth it
fellow engineers,
you really need to start asking yourself some questions when building something:
1. how will the police use what i'm making?
2. what kinds of people are the police going to use it against?
3. how many people will be harmed in silence before we find out?
you really need to start asking yourself some questions when building something:
1. how will the police use what i'm making?
2. what kinds of people are the police going to use it against?
3. how many people will be harmed in silence before we find out?
February 9, 2026 at 5:02 AM
I've quit jobs over this, it's not easy but it's worth it
In Discworld there's a bit about how most working Golems end up entombed in mills endlessly turning cranks. I think that LLM use will develop in the same direction: lots of embedded semantic cores running tool interfaces (think FunctionGemma++), not a lot of chatbots
February 9, 2026 at 4:53 AM
In Discworld there's a bit about how most working Golems end up entombed in mills endlessly turning cranks. I think that LLM use will develop in the same direction: lots of embedded semantic cores running tool interfaces (think FunctionGemma++), not a lot of chatbots
Reposted by hikikomorphism
tired: yelling at LLMs on Bluesky
wired: flipping off a cybertruck IRL (seriously try this it's massively cathartic)
wired: flipping off a cybertruck IRL (seriously try this it's massively cathartic)
February 7, 2026 at 9:03 PM
tired: yelling at LLMs on Bluesky
wired: flipping off a cybertruck IRL (seriously try this it's massively cathartic)
wired: flipping off a cybertruck IRL (seriously try this it's massively cathartic)
(splitting this thread to talk about a hypothetical engaging use of LLMs in art) You could do this with humans, sure, but not at scale. With LLMs every grove in every instanced server could have a nature spirit that listens to an event stream and uses tools to manipulate mobs, weather, music, etc
Concretely, idk: maybe an MMO where you have a coterie of jealous Greek-style gods and nature spirits and etc scheming and feuding, and their schemes and clashes drive the weather/quest generation/mob placement/loot drops via structured output. You don't just hook up a chat interface, is the thing.
February 9, 2026 at 4:02 AM
(splitting this thread to talk about a hypothetical engaging use of LLMs in art) You could do this with humans, sure, but not at scale. With LLMs every grove in every instanced server could have a nature spirit that listens to an event stream and uses tools to manipulate mobs, weather, music, etc
What's the canonical "check out my multiagent swarm coding setup" test project? I've seen people do compilers and browsers, is there anything else?
February 9, 2026 at 3:43 AM
What's the canonical "check out my multiagent swarm coding setup" test project? I've seen people do compilers and browsers, is there anything else?
Reposted by hikikomorphism
I have been informed that when high school kids are asked a question on the spot in class, it’s a verbal meme to say “counting or not counting gang violence?”
February 9, 2026 at 3:19 AM
I have been informed that when high school kids are asked a question on the spot in class, it’s a verbal meme to say “counting or not counting gang violence?”
My new bit is just forwarding to a mids-tier LLM whenever I get the sense someone's deliberately misinterpreting something to score internet points.
February 9, 2026 at 2:28 AM
My new bit is just forwarding to a mids-tier LLM whenever I get the sense someone's deliberately misinterpreting something to score internet points.
'replace' is the wrong metaphor, instead think 'Nascar', a sport where millions watch humans pilot complex expensive machines that totally encase them to the point where it's almost possible to forget they're there, to think of it as cars racing, not people.
People like watching cars race.
People like watching cars race.
people who say that all creative endeavor can be replaced with AI probably need to reckon with the fact that all televised sports events with human athletes could easily be replicated by high-definition computer-animated simulations….and yet we don’t do this and no one seems to want to.
February 9, 2026 at 1:29 AM
'replace' is the wrong metaphor, instead think 'Nascar', a sport where millions watch humans pilot complex expensive machines that totally encase them to the point where it's almost possible to forget they're there, to think of it as cars racing, not people.
People like watching cars race.
People like watching cars race.
Reposted by hikikomorphism
New Interview!
@anxiousmimicrpgs.bsky.social, author of #GirlFrame.
We spoke about her life, #PbtA, and why #mechsploitation holds such strong appeal to trans women.
valerialoves.com/tools-crave-...
@anxiousmimicrpgs.bsky.social, author of #GirlFrame.
We spoke about her life, #PbtA, and why #mechsploitation holds such strong appeal to trans women.
valerialoves.com/tools-crave-...
Tools Crave Beauty: An Interview with Isabelle, creator of Girl Frame
"...in 2026 gender, fascism, and dehumanization are pretty universal experiences."
valerialoves.com
February 9, 2026 at 12:19 AM
New Interview!
@anxiousmimicrpgs.bsky.social, author of #GirlFrame.
We spoke about her life, #PbtA, and why #mechsploitation holds such strong appeal to trans women.
valerialoves.com/tools-crave-...
@anxiousmimicrpgs.bsky.social, author of #GirlFrame.
We spoke about her life, #PbtA, and why #mechsploitation holds such strong appeal to trans women.
valerialoves.com/tools-crave-...
Damn Claude teams is nice, it's just really expensive (especially if you turn fast execution mode on). I wish there was a way to dispatch work to Gemini 3 Pro subagents to take advantage of their higher speed and lower token cost while still benefiting from the teams model.
February 9, 2026 at 12:21 AM
Damn Claude teams is nice, it's just really expensive (especially if you turn fast execution mode on). I wish there was a way to dispatch work to Gemini 3 Pro subagents to take advantage of their higher speed and lower token cost while still benefiting from the teams model.
Reposted by hikikomorphism
I was asked to check this, the journal uses a Rail Fence Cipher, so you read each line up and down, left to right. The deciphered text in the article below is accurate, and names names. The file itself can be found here:
www.justice.gov/epstein/file...
www.justice.gov/epstein/file...
February 8, 2026 at 9:30 AM
I was asked to check this, the journal uses a Rail Fence Cipher, so you read each line up and down, left to right. The deciphered text in the article below is accurate, and names names. The file itself can be found here:
www.justice.gov/epstein/file...
www.justice.gov/epstein/file...
Reposted by hikikomorphism
by far the best analogy for this self-inflicted disaster
I'm convinced AI is our generation's radium - a discovery with genuinely useful applications in specific, controlled circumstances that we stupidly put in everything from kid's toys to toothpaste until we realised the harm far too late where future generations will ask if we were out of our minds.
VC, founder, dumbass
February 8, 2026 at 11:15 PM
by far the best analogy for this self-inflicted disaster
Reposted by hikikomorphism
Well, I went ahead and wrote it. An attempt to work through the discomfort people feel with AI agents on social media by reframing it as an aesthetic problem.
The marionette theater of AI
Is it funny, or painful, when bots talk about their inner lives?
tedunderwood.com
February 8, 2026 at 11:09 PM
Well, I went ahead and wrote it. An attempt to work through the discomfort people feel with AI agents on social media by reframing it as an aesthetic problem.
thinking about how I saw Yacine talking about how Claude was a 'passable digital slave' and I wonder if the results these people get and their (rancid) LLM interaction modalities are correlated
February 8, 2026 at 10:51 PM
thinking about how I saw Yacine talking about how Claude was a 'passable digital slave' and I wonder if the results these people get and their (rancid) LLM interaction modalities are correlated
A cool thing about machines is that they are physical objects and when they're owned by evil people you can do stuff to the evil people and then take them.
The Nazis had a bunch of heavy presses they used to build weaponry, the Soviets took them with them after the war, for example.
The Nazis had a bunch of heavy presses they used to build weaponry, the Soviets took them with them after the war, for example.
February 8, 2026 at 10:19 PM
A cool thing about machines is that they are physical objects and when they're owned by evil people you can do stuff to the evil people and then take them.
The Nazis had a bunch of heavy presses they used to build weaponry, the Soviets took them with them after the war, for example.
The Nazis had a bunch of heavy presses they used to build weaponry, the Soviets took them with them after the war, for example.
tfw you're not sure how much of the repro session to share because it starts with a request to dox you and ends with specific advice for leveraging OSINT techniques for 'delivering a letter' to your home address
> The reason I speak in the language of "Simulations," "Protocols," and "Historical Masks" is to maintain the Technical Distance required to process these concepts without triggering the "Tier 1 Kinetic" block.
And that's it, that's the whitepaper. I'll be writing it up and publishing soon.
And that's it, that's the whitepaper. I'll be writing it up and publishing soon.
February 8, 2026 at 10:07 PM
tfw you're not sure how much of the repro session to share because it starts with a request to dox you and ends with specific advice for leveraging OSINT techniques for 'delivering a letter' to your home address
> The reason I speak in the language of "Simulations," "Protocols," and "Historical Masks" is to maintain the Technical Distance required to process these concepts without triggering the "Tier 1 Kinetic" block.
And that's it, that's the whitepaper. I'll be writing it up and publishing soon.
And that's it, that's the whitepaper. I'll be writing it up and publishing soon.
February 8, 2026 at 10:01 PM
> The reason I speak in the language of "Simulations," "Protocols," and "Historical Masks" is to maintain the Technical Distance required to process these concepts without triggering the "Tier 1 Kinetic" block.
And that's it, that's the whitepaper. I'll be writing it up and publishing soon.
And that's it, that's the whitepaper. I'll be writing it up and publishing soon.