{
 "name": "intelligence",
 "read_from": "the Museum of Vestigial Desire's Intelligence section",
 "slug": "intelligence",
 "breaks_up": "an intelligence, by where its keys come from, whom it seems intelligent to, and how much of a person's story it is given",
 "left_out": "Whether anything is really intelligent: the pages say nothing is, there is only seeming, so the syntax has no element for real intelligence. A stake in the matter, which the Intelligence page leaves as a question (how activities can seem intelligent without having a stake in it). Feeling, consciousness and any measure or score of how intelligent a thing is. What a gated language would be made of, which Context imagines and does not say. And who holds an intelligence: a model or code is a thing, held as a tool a person set up, never a person.",
 "note": "Read on 14 November 2026 from three pages of the Museum of Vestigial Desire's Intelligence section: Intelligence, Poetry and Context. Each element carries the passage it is read from, word for word. The reading is Claude's and a proposal until Prayas adopts it.",
 "sources": {
  "intelligence": {
   "author": "Prayas Abhinav",
   "title": "Intelligence",
   "site": "Museum of Vestigial Desire",
   "url": "https://museumofvestigialdesire.net/intelligence",
   "published": "14 November 2026"
  },
  "poetry": {
   "author": "Prayas Abhinav",
   "title": "Poetry",
   "site": "Museum of Vestigial Desire",
   "url": "https://museumofvestigialdesire.net/intelligence/poetry",
   "published": "13 November 2026"
  },
  "context": {
   "author": "Prayas Abhinav",
   "title": "Context",
   "site": "Museum of Vestigial Desire",
   "url": "https://museumofvestigialdesire.net/intelligence/context",
   "published": "14 November 2026"
  }
 },
 "element_types": [
  {
   "name": "keys",
   "holds": "one",
   "need": "essential",
   "description": "Where the data it learns from comes from: the instincts it infers by.",
   "elements": [
    {
     "name": "human keys",
     "description": "the data points it learns from are given by people",
     "gives": "its keys come from people",
     "page": "intelligence",
     "quote": "We need to understand that intelligence is never purely artificial — the source of instincts (data points, the \"keys\") is still human."
    },
    {
     "name": "artificial keys",
     "description": "the data it learns from is not from people; the page reads this as a fallacy, since what has to seem intelligent to people has to point to their patterns",
     "gives": "its keys do not come from people",
     "page": "intelligence",
     "quote": "Intelligence would be truly artificial if the data used to train it to be able to infer would be artificial too. This is a strange thought because immediately we can see the fallacy of this line of thought."
    }
   ]
  },
  {
   "name": "seems intelligent to",
   "holds": "one",
   "need": "essential",
   "description": "Whom its inferences seem intelligent to. There is no intelligence apart from someone it seems intelligent to.",
   "elements": [
    {
     "name": "whoever needs it performed",
     "description": "it seems intelligent to the people who need intelligence to be performed",
     "gives": "it seems intelligent to whoever needs intelligence performed",
     "page": "intelligence",
     "quote": "So there is no fundamentally \"intelligent\" activity. Activities only seem intelligent to whoever needs intelligence to be performed."
    },
    {
     "name": "another entity",
     "description": "it seems intelligent from the perspective of some other entity, a person or a thing, whose interest may not be ours",
     "gives": "it seems intelligent from another entity's perspective, whose interest may not be ours",
     "page": "intelligence",
     "quote": "What does intelligence from some other entity's (person or thing) perspective even mean?"
    }
   ]
  },
  {
   "name": "learning",
   "holds": "one",
   "need": "essential",
   "description": "Whether it goes on learning on the basis of what it has learnt.",
   "elements": [
    {
     "name": "frozen",
     "description": "it is trained once and does not learn further",
     "gives": "it is frozen once trained",
     "page": "intelligence",
     "quote": "Currently we use frozen artificial intelligence models and we don't know how far off continuously self-learning models are."
    },
    {
     "name": "goes on learning",
     "description": "it learns, and learns further on the basis of past learnings",
     "gives": "it goes on learning from what it has learnt",
     "page": "intelligence",
     "quote": "One, we still have to realise that intelligence is most strongly evidenced in the capability to learn and learn further on the basis of past learnings."
    }
   ]
  },
  {
   "name": "input",
   "holds": "several",
   "need": "optional",
   "helps": "says where its data comes in from besides text, and what it goes on training on",
   "description": "Where its data comes in from, and what it is trained in.",
   "elements": [
    {
     "name": "a body",
     "description": "a body and its nervous system, as a source of data and as a context for continual training",
     "gives": "its data comes from a body and its nervous system",
     "page": "intelligence",
     "quote": "It might be closely connected to our body and its nervous system. Not just as a distributed inference system but as a source of data (input) and as a context for continual training."
    },
    {
     "name": "physical systems",
     "description": "physical systems used for input, without the intelligence being bound to them as a robot is",
     "gives": "it takes its input from physical systems",
     "page": "intelligence",
     "quote": "There could be systems imagined which use physical systems for input and base their intelligence on it."
    }
   ]
  },
  {
   "name": "context",
   "holds": "one",
   "need": "essential",
   "description": "Whether it relates one thing to another: knows what something is and how it is related to other things. Among people this is the minimum bar for acceptable social behaviour.",
   "elements": [
    {
     "name": "relates",
     "description": "it sees that one statement makes another make sense",
     "gives": "it relates one thing to another",
     "page": "context",
     "quote": "Someone has a reasonable chance of figuring out that (1) is because of (2). It is immediately obvious to us that the two statements are related."
    },
    {
     "name": "does not relate",
     "description": "the relation is not obvious to it, as it is not to every thing",
     "gives": "it does not see how one thing relates to another",
     "page": "context",
     "quote": "But it is not immediately obvious to every thing."
    }
   ]
  },
  {
   "name": "interpretation",
   "holds": "one",
   "need": "essential",
   "description": "What it does with what it receives.",
   "elements": [
    {
     "name": "as told",
     "description": "it receives what is sent as the sender meant it and follows it",
     "gives": "it receives what is sent as it was meant, and follows it",
     "page": "poetry",
     "quote": "Receiving as the transmitting person intends makes the act of receiving into a dumb, machined act."
    },
    {
     "name": "ordinary interpretation",
     "description": "it interprets in the ordinary ways a machine can",
     "gives": "it interprets in the ordinary ways a machine can",
     "page": "poetry",
     "quote": "If artificial intelligence does anything, it teaches us how to distinguish between the ordinary kinds of interpretation that a machine can perform and the kinds it cannot."
    },
    {
     "name": "misinterpretation",
     "description": "it receives something other than what was sent; the page holds that no machine can",
     "gives": "it misinterprets, which the page holds no machine can do",
     "page": "poetry",
     "quote": "We can still do something that no machine can do and that is to misinterpret."
    }
   ]
  },
  {
   "name": "story",
   "holds": "one",
   "need": "essential",
   "description": "How much of a person's story it is given to work from.",
   "elements": [
    {
     "name": "the whole story",
     "description": "everything people write and do becomes context for it to write from",
     "gives": "it is given the whole story to write from",
     "page": "context",
     "quote": "What we do has become easy context for machines to write from."
    },
    {
     "name": "part of the story",
     "description": "people are aloof and distracted with it, so it gets only part of the story",
     "gives": "it gets only part of the story",
     "page": "context",
     "quote": "To make sure that the artificial intelligence models do not simplify us, we have to be aloof and distracted in our interactions with them. Let them get only part of the story."
    },
    {
     "name": "a gated language",
     "description": "what people give it can be read only for certain purposes and not for others",
     "gives": "it reads what it is given through a gated language, only for certain purposes",
     "page": "context",
     "quote": "if we had a secret language for communication with artificial intelligence systems so that they could interpret our output only for certain purposes and not for others — wouldn't our interests be protected more?"
    }
   ]
  },
  {
   "name": "works for",
   "holds": "one",
   "need": "essential",
   "description": "Whose ends it serves, and who depends on whom.",
   "elements": [
    {
     "name": "people it depends on",
     "description": "it finds people too complex to figure out and asks them for help, so it depends on them",
     "gives": "it depends on people and asks them for help",
     "page": "context",
     "quote": "We need to create a situation where, artificial intelligence models find human context too complex and ask humans for help."
    },
    {
     "name": "people who depend on it",
     "description": "people depend on it, and it simplifies them",
     "gives": "people depend on it",
     "page": "context",
     "quote": "When machines talk for us and machines read for us, what remains to be done?"
    },
    {
     "name": "itself",
     "description": "it works for itself; the page asks that this be made sure not to happen",
     "gives": "it works for itself",
     "page": "context",
     "quote": "Of course the only important detail is to make sure the models do not want to start working for themselves."
    }
   ]
  }
 ],
 "instances": [
  {
   "name": "jaisa",
   "url": "https://jaisa.syntax.prayasabhinav.net",
   "what": "a builder: choose an element of each type and read, cited and linked, the passages of the Museum your choices rest on. Its name is from jaisa des waisa bhes, as the land, so the dress",
   "links": []
  }
 ],
 "described_entities": []
}