Curt Doolittle

Form: Project Update

  • We’ve done the work. At present it’s eight volumes of rigorous work. We will beg

    We’ve done the work. At present it’s eight volumes of rigorous work. We will begin publishing them incrementally this year.

    If you have particular ideas that then we would love to hear them. However, the majority of conservative thought presumes an equality of instinct, intuition, bias, ability, and interest that does not exist, and as such policies cannot be stated under the pretense that people will behave as desired.
    Instead we must govern with the human beings that exist and will exist, who are not bank slates, and relatively immutable, especially without education and continuous social and institutional enforcement.
    And as such those legal and organizational institutions we produce must maintain incentives independently of the biases of the people who occupy them.

    Source date (UTC): 2025-10-21 18:52:42 UTC

    Original post: https://twitter.com/i/web/status/1980708842905432467

  • In volume one – the crisis of the age, we survey the cycles and add to them, and

    In volume one – the crisis of the age, we survey the cycles and add to them, and unlike others, we explain why they occur and what to do about them. Given that quigley was an influence, and was relatively recent, the similarity is clear.

    Source date (UTC): 2025-10-21 03:58:34 UTC

    Original post: https://twitter.com/i/web/status/1980483826133659968

  • Website is still in early draft form. Ariella is trying to get the content to me

    Website is still in early draft form. Ariella is trying to get the content to me this week. But you can see it at runcible dot com.

    We made a pitch last week and it went flawlessly, although we were missing the three-to-six key slides that we feel are best argument and explanation.

    So we feel we’ll be ready at the end of this week to scale our outreach.

    About two months later than we’d hoped, but then, although the value of what we’re selling is obvious, it’s not like what we’re selling is easy to explain. If it was it wouldn’t have taken me 25 years, three companies, and you and your PhD to make it all happen.

    Source date (UTC): 2025-10-12 19:28:21 UTC

    Original post: https://twitter.com/i/web/status/1977456322477199420

  • We are maintaining the separation between NLI as the legal research and activism

    We are maintaining the separation between NLI as the legal research and activism organization, and Runcible as the applied technology organization. Thus we de-politicize runcible while retaining the activism of NLI.

    As such we are merging Reality by Chanting (RBC) into Runcible by repatriating RBC assets from Europe (UK/UA) into the USA and Runcible. And Licensing the research from NLI to Runcible. NLI is prohibited from allowing what’s called ‘capture’ of the IP, and as such converting NLI work from truth, reciprocity, possibility, natural law, and what we call ‘perfect government’ into some corruption of it.

    This integration of NLI IP, with RBC IP, into Runcible, creates two business lines in Runcible, meaning the Runcible Intelligence layer which we license, and the Runcible application front end and Runcible AI backend as a commercial application for running organizations of any scale from personal to business to industrial to government to military.

    Even today we can see how Brandon Hayes (president of NLI) is conducting reform (lawfare) against the proceduralism and bureaucracy endemic to governments, while I and some others are driving Runcible to commercialization. That commercialization will allow revenue via licensing from NLI to RBC to pursue both socio-political and commercial goals.

    The end result w hope is saturation of the population with an information system on one hand and tha revolution in desktop and laptop interfaces what synthesize AI and Business an Government.

    This two pronged approach, we assume, will do more to indirectly educate the public and transform public discourse, than any of our attempts to modify the law.

    Only then will the law be modifiable.

    Only then can we prevent a civil war that is growing from warm to hot faster than we can act.

    Curt Doolittle
    NLI and Runcible.

    Source date (UTC): 2025-10-12 19:16:26 UTC

    Original post: https://twitter.com/i/web/status/1977453324946505785

  • Simple but I have to write it up. I’ll try to do so over the following week, eve

    Simple but I have to write it up. I’ll try to do so over the following week, even though I have a lot going on this week.

    Source date (UTC): 2025-10-06 22:58:17 UTC

    Original post: https://twitter.com/i/web/status/1975334826036359617

  • Yesterday I worked on the masculine-feminine macro cycle in civilizations and I

    Yesterday I worked on the masculine-feminine macro cycle in civilizations and I think it ‘completes’ the civilizational cycle causes – and explains why we are where we our in our civilizational decline … in no small part by the inclusion of women into education, administration, and politics.

    Source date (UTC): 2025-10-06 16:32:39 UTC

    Original post: https://twitter.com/i/web/status/1975237778410762692

  • (Runcible) We have addressed the following features over the past ten days or so

    (Runcible)
    We have addressed the following features over the past ten days or so:
    1. Implementing: Enumeration, Operationalization, serialization, adversarial disambiguation (EOSA) – Disambiguation methodology.
    2. Forcing Glossary members (terms) as immutable measures.
    3.

    Source date (UTC): 2025-09-26 01:33:16 UTC

    Original post: https://twitter.com/i/web/status/1971387562133762309

  • (NLI: Runcible): What not to do (What we’ve learned) – Don’t try to “pretrain on

    (NLI: Runcible):
    What not to do (What we’ve learned)
    – Don’t try to “pretrain on the books” hoping the model memorizes them. You lose provenance and get drift. Use RAG + span-level supervision instead.
    – Don’t ship free-form answers. Make the schema the product.

    Source date (UTC): 2025-09-23 15:43:38 UTC

    Original post: https://twitter.com/i/web/status/1970514401372639552

  • Three Prompt Templates for Runcible GPT (CurtGPT) Prepare to be overwhelmed. The

    Three Prompt Templates for Runcible GPT (CurtGPT)

    Prepare to be overwhelmed. These prompts expose the entire process that we use for testing the truth and reciprocity of claims and assertions. The examples we use are simple but you can put a news article or a policy doc into it and get a terrifyingly accurate result.

    I have not incorporated our science of lying into the model yet – partly because I”m a bit afraid of what it will demonstrate – the pervasiveness of lying in our daily lives. But that will come in the future.

    Proceed step-by-step. Show PLAN → OPS → AUDIT → 11-STEP.
    –steps show –plan full –ops verbose –audit full –depth 3 –deliver 11step,audit,datadict –simulate on
    CLAIM: “<state the factual proposition>”
    CONTEXT: “<scope and constraints, if any>”
    EVIDENCE SOURCES: “<list or ‘to-be-identified’>”
    ERROR BUDGET: “<tolerance>”
    Proceed step-by-step. Show PLAN → OPS → AUDIT → 11-STEP.
    –steps show –plan full –ops verbose –audit full –depth 3 –exceptions warranted –deliver 11step,audit,checklist –simulate on
    TASK: “Evaluate policy X for institutionalized reciprocity and computable enforcement.”
    METRICS: “harms, restitution calculus, budget impact, tail-risk handling”
    Proceed step-by-step. Show PLAN → OPS → AUDIT → 11-STEP.
    –steps show –plan full –ops verbose –audit full –depth 3 –deliver 11step,audit,datadict,checklist
    TASK: “Specify a decision procedure for [domain].”
    CONSTRAINT: “No discretionary thresholds; exceptions require performer warranty.”

    Source date (UTC): 2025-09-15 20:11:39 UTC

    Original post: https://x.com/i/articles/1967682746626818285

  • (NLI – RUNCIBLE) GOOD STUFF: I’ve added both volume 0 – the history of man, and

    (NLI – RUNCIBLE)
    GOOD STUFF: I’ve added both volume 0 – the history of man, and an appendix: the abrahamic method of deception to the Runcible demo. That’s the good part.

    BAD STUFF: the corpus, while still incomplete, is large enough that Openai’s GPT5 hallucinates so badly that it’s almost unusable.

    This means that we MUST train a model in order to demonstrate the work. We cannot rely on Prompts and RAG.

    This isn’t good. It’s a huge time and cost sink.

    Source date (UTC): 2025-09-15 17:35:16 UTC

    Original post: https://twitter.com/i/web/status/1967643392118284571