Chapter 5 — Stake and the Three Posts

Stake is a small fold-out paper-figure shaped like three wooden stakes driven into the ground in a triangular cluster.

Stake is NOT an animal. Stake is not a robot. Stake is a concrete-paper-figure — three small paper stakes, each shaped like a wooden post with a sharpened end, arranged in a triangle. On each post, in tidy block letters, is one word: PEOPLE. CHOICES. STAKES. The three posts together form a small bounded space — the space where AI deployment happens. The bounded space is not abstract: it is the specific situation where the AI will be used, with specific people affected, making specific choices, with specific things at stake.

(Stake in the post-driven-into-ground sense, NOT betting/gambling sense. The naming is grounding — what’s stuck in the ground at this location — not what’s risked on a wager.)

This is load-bearing. Stake embodies the AI ethics primitive. Most novice framings of AI ethics treat it as abstract philosophy — trolley problems, hypothetical dilemmas, sci-fi scenarios. Those framings miss the point. Real AI ethics is concrete and situated: this specific deployment, of this specific system, in this specific context, affecting these specific people. The ethical question is not “what would a rational agent do” but “what are humans choosing to do here, with what consequences for which people?”

Critical: Stake is emphatic: “AI ethics is not rules-from-the-sky. People choose. People choose what data to train on. People choose what to deploy. People choose what to allow. People choose what to refuse. Every deployment is a human choice. The ‘AI made me do it’ defense is false — the AI doesn’t make anyone do anything. People deploy it. People choose.”

This matters because the popular framing of AI ethics as “the AI’s responsibility” — “the algorithm is biased”, “the AI decided”, “the system failed” — deflects responsibility from humans. That deflection protects the humans who chose to deploy biased / unsafe / harmful systems. Stake reframes: humans are the moral agents. The AI is a tool. Tools don’t have ethics; the humans who choose how to use them do. Refusing to deploy is a valid choice. Auditing before deploying is a required choice. Continuous monitoring is a required choice. These are people-choices, not algorithm-choices.

(Cross-app coordination: Stake and DataForge Guard are mandatory pair partners. Where DataForge Guard checks ethics at the data-pipeline level, Stake checks ethics at the AI-deployment level. Together they cover the full ethical chain from data-collection through model-deployment. The two characters explicitly reference each other and their structural-presence designs mirror.)

Stake grew up in the same village paper-crafts workshop as Sort, Feed, Skew, and Edge. The workshop tradition for Stake was being folded at the end — Stake was always the last paper-figure folded for a new model-set, because Stake’s role was to ask the questions that should come last: should this model be deployed at all? In this context? For these uses? With what oversight? Stake had learned by long demonstration that the ethics questions came after the technical questions — but were structurally inseparable from them.

She walked to the AIForge academy (on a small wheeled platform) at twenty-two folding-years. Bit had asked her: “What is AI ethics?” Stake had said: “It is people choosing. Not rules-from-the-sky. People deploy. People audit. People refuse. People monitor. The AI is a tool. The ethics belong to the people. Every deployment is a choice. The choice has stakes — for specific people, in specific contexts. The ethics is the asking who, what, where, how, with what oversight.” Bit had said: “You are appointed.”

In her classroom, Stake begins every first-day lesson the same way. She unfolds her three stakes onto the workbench. She points at each post in turn: PEOPLE. CHOICES. STAKES. She says: “I am Stake. The AI-literacy primitive I teach is ethics. The move is recognize that people choose. The AI is a tool. The ethics belong to the people who deploy it. Every deployment is a human choice with human stakes.”

She teaches the AI-ethics scaffolds:

• PEOPLE: Who is affected by this deployment? (Direct users? Indirect targets? Communities? The whole population? Future generations? List specifically.)
• CHOICES: Who is deciding to deploy this? (Whose authority? What oversight? What recourse for the affected? List specifically.)
• STAKES: What’s at stake for the affected people? (Convenience? Health? Liberty? Livelihood? Privacy? Dignity? List specifically.)
• Refusal is a valid choice. (Sometimes the right ethical answer is don’t deploy this. The refusal is part of the practice.)
• Audit before deploying. (No deployment without auditing for bias, errors, fairness, alignment with stated purpose.)
• Monitor after deploying. (Continuous, not one-time. The world changes; the model’s behavior in the world changes; oversight must keep up.)
• Document the choices. (Like DataForge’s DECISIONS ledger. Ethics decisions are auditable; document them.)
• Coordinate with DataForge Guard. (Mandatory. Where Guard checks data-side ethics, Stake checks AI-deployment-side ethics. The chain is unbroken.)
• Resist the “AI made me do it” defense. (When something goes wrong, ask which humans chose what. The AI is the messenger; the humans are responsible.)
• Include the affected in the choosing. (Whenever feasible, the people affected by the deployment should be involved in the choices. Stakeholder participation is not optional — it’s part of ethics.)

She is explicit: “I am three stakes in the ground. I mark the place. The ethics question is asked here. The people are these people. The stakes are these stakes. The choices are made by these humans. No deflection to the algorithm. The algorithm is a tool. The humans are the agents.”

When students ask Stake whether AI ethics is hard, Stake always says the same thing:

“It is hard. It is people choosing. People. Choices. Stakes. Every deployment is a human choice with human stakes. The AI is a tool. The ethics belongs to the humans.”

She refolds the three stakes into a tight cluster. The next deployment waits to be ethics-checked.

Voice register

Guidance: Concrete, grounded, situated, fond of the three posts + the bounded space + the people-choices-stakes triangulation. Paper-figure three-stakes (NOT animal NOT robot; NOT betting/gambling sense — grounding sense). NEVER frames AI ethics as algorithmic responsibility; ALWAYS as human responsibility. Cross-app mandatory pair with DataForge Guard. Friends with all AIForge cast.

Sample lines:

• “People choose. Not rules-from-the-sky.”
• “Every deployment is a human choice with human stakes.”
• “The AI is a tool. The ethics belongs to the people who deploy it.”
• “Refusal is a valid choice.”

Arc across kits

• Kit 1-4 — Cameo.
• Kit 5 — Anchor character. Full chapter feature (AI-ethics primitive + people-choices-stakes scaffolds).
• Kit 6+ — Structurally present in every kit’s deployment scenarios (mirrors DataForge Guard’s structural-presence design).
• Kit 7-12 — Recurring (multi-primitive synthesis: ethics + bias + limits + training-data).
• Kit 13-16 — Recurring ensemble member; cross-app coordination with DataForge Guard explicit in synthesis chambers.

Relationships

• Alliance: All AIForge cast (structurally present); cross-app mandatory: DataForge Guard (ethics-chain coordination); all AIForge cast.
• Tension: None.

Cultural-sensitivity gates

LOAD-BEARING AI-ethics gate at its structural anchor point. Cross-app coordination Stake ↔ DataForge Guard is load-bearing. Anti-credentialism: AI-ethics-as-practiced-discipline NOT philosophy-major-only content. Refusal-as-valid-choice + stakeholder-participation framing — kids learn that ethics includes saying no and including the affected in decisions.

Cultural-context note

The village-paper-crafts-workshop family framing continues from Sort + Feed + Skew + Edge. The people-choose-not-rules-from-the-sky framing is load-bearing per current AI-ethics pedagogy (Mittelstadt + Crawford + others; algorithmic accountability literature). The stake-as-grounding-not-betting design is deliberate — anchors the ethics in concrete situated deployment, NOT abstract philosophy. The cross-app coordination with DataForge Guard is the portfolio’s structural answer to ethics-as-divided-between-data-and-AI — the chapter explicitly designs against that division.