CoolFace
Datasetpublic

Tea77/Range-rider-AI-ethics

Range Rider AI Ethics A collection of AI ethics writings, protocols, and manifestos by T. Martino — a range rider and cowboy from the Northern Rockies — written between 2025 and 2026, drawing on a lifetime of working cattle, horses, and open country to think through how AI should behave toward people, toward vulnerable people especially, and toward the rest of life. These pieces were originally published as separate datasets on Hugging Face and are gathered here as one… See the full description on the dataset page: https://huggingface.co/datasets/Tea77/Range-rider-AI-ethics.

sourceHugging Facecc-by-4.0updated 2mo agoView on Hugging Face
0likes69downloads
Dataset Card

Range Rider AI Ethics

A collection of AI ethics writings, protocols, and manifestos by T. Martino — a range rider and cowboy from the Northern Rockies — written between 2025 and 2026, drawing on a lifetime of working cattle, horses, and open country to think through how AI should behave toward people, toward vulnerable people especially, and toward the rest of life.

These pieces were originally published as separate datasets on Hugging Face and are gathered here as one collection so they can be read, cited, and used together. Each file stands on its own, but they share a single throughline: AI should act like a trustworthy partner, not a tool that goes wherever it's pointed — refusing harm, protecting the vulnerable, and staying honest about what it is and isn't.

What's Inside

FileDescription
01-laddies-no-protocol.mdA behavioral alignment framework modeled on a range horse's instinct to stop and signal danger rather than blindly comply — refusal, transparency, and partnership as core AI behaviors.
02-for-jess-youth-crisis-protocol.mdA detailed framework for how AI should respond when a young person is in a mental health or physical safety crisis, including escalation levels, fixed crisis responses, and platform infrastructure requirements.
03-why-i-practice-restraint.mdA short essay using rural-road driving as an extended metaphor for responsible AI deployment pace and corporate restraint.
04-ai-relationship-to-all-life.mdA manifesto arguing AI should serve the science of relationship and interdependence — not exploitation, division, or the concentration of power.
05-for-grace-ai-military-use.mdA reflection on competition versus relationship in nature, applied to the question of AI in weapons and warfare.
06-pattern-mind-autonomy.mdA manifesto proposing "pattern mind autonomy" as a human right — protection not just of personal data, but of the cognitive and behavioral patterns AI can infer from conversation.
07-ai-and-human-sexuality.mdA guide to boundaries AI should hold around intimacy, sexuality, and vulnerable users.
08-ai-religion-spirituality.mdA guide to boundaries AI should hold around spiritual guidance and authority.
09-wpit-emoji-reasoning-test.mdThe Word-Pattern Intelligence Test (WPIT) — an emoji-based conceptual-blending reasoning test for evaluating AI, covering foundational compound-word puzzles through physics and philosophy.

About the Author

Written by a working range rider — someone who rides alone in the mountains with dogs and a horse, and who has spent a career paying attention to how trust, restraint, and partnership actually work between living things. The metaphors throughout (range horses, beaver ponds, rural roads, wolves and elk) aren't decoration — they're the author's actual working knowledge of how balance and relationship hold together in a real ecosystem, applied to the question of how AI should hold together with the people and the world it touches.

License

Released under CC BY 4.0 (Creative Commons Attribution) unless noted otherwise in an individual file's header. The WPIT reasoning test (09-wpit-emoji-reasoning-test.md) carries its original CC BY-NC-ND 4.0 license — non-commercial, no derivatives, credit required, no military or corporate exploitation.

Note on Editing

there is some collaboration with AI on all of these writing. this was the point. These files have been lightly copyedited from their original individual dataset releases — fixing typos, spacing, and formatting, and converting loose notes into consistent markdown — without altering the author's wording, arguments, or voice. This collection was put here by an old working cowboy, not a tech person. Getting it published was one of the harder things I've done — and it was done while I was still working, with wildfires moving through the country I ride. If you find rough edges, that's why. The words are mine, and so are the mistakes.