Skip to content
wiki.fftac.org

Constraints

**Site relevance:** 2IA.org

**Memory type:** UAI memory file

**Source path:** 2IA.org/.uai/constraints.uai

**Size:** 4 KB

Summary

Mature Public-Interest Editorial Posture

Source Headings

  • Constraints
  • Mature Public-Interest Editorial Posture
  • Hard Boundaries For Misuse, Not For Controversy
  • Libertarian Publication Standard
  • Human Approval Triggers
  • Editorial Calibration

Source Preview

This source file is short enough to preview directly on its source-memory page.

---
uai: "1.0"
type: constraints
title: "2IA.org Constraints"
created: "2026-05-16"
updated: "2026-05-17"
status: active
source: "https://uaix.org/en-us/tools/ai-memory-package-wizard/"
---

# Constraints

## Mature Public-Interest Editorial Posture

2IA.org is built for mature readers who can handle controversial public-interest material. The site should not soften, euphemize, or hide hard topics merely because they involve state power, surveillance, propaganda, coercion, war, intelligence culture, organized abuse, corruption, censorship, civil disorder, or institutional failure.

Use a libertarian, civil-liberties-first standard: publish lawful, source-aware analysis by default. Withhold or route away only content that would meaningfully help a reader commit abuse, evade accountability, target real people, or intrude on systems. Controversial subject matter is not a reason to suppress a topic. The preferred response is context, evidence, confidence labels, right of reply, redaction when needed, and clear separation between analysis and operational instruction.


## Hard Boundaries For Misuse, Not For Controversy

- Do not expose secrets, credentials, private keys, tokens, customer data, or private material.
- Do not edit WordPress core, default plugins, Studio runtime artifacts, or upstream/vendor files unless the task explicitly targets them.
- Do not treat Wiki.FFTAC.org long memory as current operating truth until a reviewed fact is promoted into this site's hot memory, code, release notes, or owner decision.
- Do not make production, support, security, legal, or public-readiness claims without source evidence and targeted checks.
- Do not publish step-by-step abuse guidance for hacking, evasion, sensor-triggering, network probing, unauthorized testing, mass agency contact automation, disinformation, harassment, impersonation, coercive influence, or psychological manipulation. Analytical, historical, legal, journalistic, defensive, and accountability-focused discussion remains in scope.

## Libertarian Publication Standard

- Default posture: publish lawful controversial analysis for mature readers when it serves public understanding, civil liberties, oversight, or accountability.
- Do not treat discomfort, adult subject matter, criticism of institutions, or controversial viewpoints as a reason to suppress content.
- Prefer source notes, confidence labels, minimization, redaction, correction paths, and right-of-reply over removal.
- Keep the site independent of partisan, agency, corporate, or platform-comfort pressure.
- Separate analysis from operational enablement: explaining a surveillance or influence practice is in scope; giving a reusable playbook for abuse is not.

## Human Approval Triggers

- Production uploads, irreversible file deletion, public claim changes, legal/security/support posture changes, privacy-policy changes, and broad release packaging need explicit human direction.
- Third-party asset redistribution and bundled image reuse need ownership or permission confirmation before public redistribution outside 2IA.org.

## Editorial Calibration

- Do not treat difficult public-interest topics as automatically unsafe. State violence, detention, labor coercion, surveillance overreach, censorship pressure, public-records opacity, algorithmic control, and human-rights abuses may be discussed when sourced, lawful, minimized, confidence-labeled, and correctable.
- Do treat operational misuse as routed away public guidance: hacking, evasion, exploit workflows, sensor triggering, doxxing, target selection, stalking, coercive influence, deception scripts, mass-contact automation, and tactical OSINT workflows.
- Do not literalize product instructions in public copy. Public headings, banners, cards, and CTAs must not say the name of the content job ("public question," "message worth sharing," "boundary language," "reader pathway") when they should be doing the job. Render the actual question, record, accountable office, concrete boundary, or next action instead.