Developer Briefing Document — For Internal & Contributor Use

Declare AI Open Standard for
AI Transparency

A universal disclosure layer for the web. Any piece of content — article, video, codebase, image, audio — carries a machine-readable and human-readable declaration of how AI contributed to it. Think Creative Commons, but for intelligence provenance.

Open Source · MIT
Free Forever
Community Governed
No Ads · No Tracking

01 — Vision

Why the web needs this now

As AI becomes embedded in the production of virtually all digital content, audiences, researchers, educators, and regulators have no standardized way to understand the provenance of what they read, watch, or hear. Declare AI solves this with a lightweight, embeddable standard.

The analogy: Nutrition labels didn't exist until regulators and consumers demanded them. AI contribution labels don't exist yet — but the moment is coming. Being the open standard that arrives first, free, and well-designed is the entire opportunity.
🏫
Education
Teachers and institutions need to know whether student work, textbooks, or research materials were AI-generated. A standard label removes the guesswork and enables policy.
High Priority
🔬
Research & Science
Journals and preprint servers need provenance. AI-assisted literature reviews, data synthesis, and figure generation need declaration to preserve scientific integrity.
High Priority
📰
Journalism & Media
News organizations using AI for summarization, translation, or image creation should disclose it. Readers deserve to know. The dispute forum catches violations.
Key Sector
🎨
Art & Creative
Artists using AI tools can declare their process transparently, protect their human creative contributions, and participate in an emerging discourse about authorship.
Cultural Layer
⚖️
Legal & Regulatory
As AI disclosure laws emerge globally (EU AI Act, etc.), Declare AI can serve as the compliance layer that platforms adopt rather than building their own.
Strategic
💊
Medicine & Health
AI-authored health content carries liability risk. Declaration + debunk scoring enables patients and professionals to evaluate the trustworthiness of what they read.
Critical

02 — System Architecture

How the pieces connect

The system has four distinct layers: the open schema standard, the embeddable widget, the backend API & dashboard, and the community forum. Each is independently useful; together they form the full platform.

Consumers
BrowserExtension (Chrome/FF/Safari)
EmbeddedJS Widget (1 script tag)
Directdeclare-ai.com Dashboard
3rd PartyCMS Plugins (WP, Ghost...)
API Layer
PublicREST API v1
AuthJWT / OAuth2
Rate LimitAnonymous · Registered
CDNJSON Delivery (Cloudflare)
Services
CoreDeclaration Service
ScoringDebunk + Deepfake Engine
SearchContent Index
NotifDispute Alerts
WorkerAI Tool Auto-Detection
Data
Primary DBPostgreSQL
CacheRedis
SearchMeilisearch
FilesS3-compatible Store
ForumDiscourse (separate)
Infra
HostFly.io / Railway (MVP)
DNS+CDNCloudflare
CI/CDGitHub Actions
MonitorGrafana + Sentry

03 — The Open Standard

declare-ai.json schema v1.0

The schema is the heart of the standard. Any creator, platform, or AI tool can generate this file. It is self-hostable — no dependency on our API required. The API adds verification, search, and dispute resolution on top.

Writing / Language
Code
Visual / Image
Audio / Music
Research / Data
Human
declare-ai.json — Full Schema Reference
{
  "$schema": "https://declare-ai.org/schema/v1.json",
  "version": "1.0.0",

  // ── Content Identification ──────────────────────
  "content_title": "AI in Medicine 2025",
  "content_url": "https://example.com/ai-medicine-2025",
  "content_type": "article",  // article | video | audio | image | code | dataset
  "content_hash": "sha256:8f4a2b...",  // optional integrity check

  // ── Declaration Metadata ────────────────────────
  "declared_by": "Dr. J. Rivera",
  "declared_by_handle": "@jrivera",
  "declared_at": "2025-09-14T10:30:00Z",
  "declaration_method": "self-declared",  // self-declared | platform-verified | community-verified

  // ── AI Contributions ────────────────────────────
  "contributions": [
    {
      "type": "writing",
      "pct": 55,
      "tool": "Claude 3.7 Sonnet",
      "tool_version": "claude-sonnet-3-7",
      "provider": "Anthropic",
      "role": "First draft, outline, editing",
      "human_hours": 1.5
    },
    {
      "type": "research",
      "pct": 18,
      "tool": "Perplexity Pro",
      "role": "Literature review, citation gathering",
      "human_hours": 2.0
    },
    {
      "type": "visual",
      "pct": 12,
      "tool": "Midjourney v7",
      "role": "Header image, diagrams",
      "human_hours": 0.5
    },
    {
      "type": "human",
      "pct": 15,
      "role": "Expert review, fact-checking, final edit",
      "human_hours": 3.0
    }
  ],

  // ── Human Contributors ──────────────────────────
  "contributors": [
    { "name": "Dr. J. Rivera", "role": "Author", "orcid": "0000-0002-1825-0097" },
    { "name": "M. Okonkwo", "role": "Editor" }
  ],

  // ── Verification & Trust ────────────────────────
  "debunk_score": 87,           // 0–100, null if unscored
  "deepfake_probability": 0.02,   // 0.0–1.0, null if not applicable
  "verified_by": "declare-ai.org",
  "verification_sig": "eyJhbGci...",  // JWT signature from API

  // ── Disputes ────────────────────────────────────
  "dispute_count": 1,
  "dispute_url": "https://forum.declare-ai.org/t/ai-medicine-2025/4421",
  "dispute_status": "open",      // open | resolved | dismissed

  // ── Legal ───────────────────────────────────────
  "license": "CC-BY-4.0",
  "jurisdiction": "US"
}
Self-hostable by design: A creator can drop declare-ai.json alongside any file, link to it from a <meta> tag, and the browser extension will auto-detect it — no API account needed. The API and dashboard add verification, search, and dispute resolution as an optional upgrade layer.

04 — Backend Services

Self-declaration portal & API

The backend has three jobs: accept and store declarations, expose them via API for the widget and extension, and run verification / scoring services. It must be horizontally scalable and 100% open source so the community can self-host.

✍️
Self-Declaration Portal
A clean web form at declare-ai.org/declare. Creator enters their content URL, fills out contribution sliders per type and tool, adds contributor names, and generates a signed JSON file they can download or host via our CDN. No account required for basic use; account required for API key, dispute tracking, and verification badge.
Core Feature
🔑
REST API
Public endpoints: GET /v1/declarations/{url} returns declaration for a URL. POST /v1/declarations submits a new one. GET /v1/search searches by tool, contributor, content type. Anonymous rate-limited; authenticated for higher limits. Fully documented via OpenAPI spec.
Developer-Facing
🧠
Scoring Engine
Two scoring services run asynchronously after declaration submission. Debunk Score uses a combination of fact-checking APIs (ClaimBuster, Google Fact Check Tools) and community votes. Deepfake Probability uses open-source models (Hive, FakeCatcher derivatives) on linked video/audio. Both are optional fields populated after async processing; creators are notified by email.
Phase 2
⚠️
Dispute System
Anyone can flag a declaration as inaccurate. A flag creates a Discourse forum thread (via Discourse API), increments dispute_count, and notifies the original declarer. Community moderators + the original creator can resolve or dismiss. Resolved disputes update the badge color in the widget from yellow to green or red.
Community Critical
MVP scope: For launch, ship the declaration form + CDN hosting + REST API + browser extension. Scoring engines and deepfake detection are Phase 2. Don't let perfect block shipped.

05 — Community Forum

Dispute & discussion layer

The forum is built on self-hosted Discourse (open source, free to run). The Declare AI API creates threads automatically when disputes are filed. Here's what a dispute thread looks like:

⚠️ Dispute: "AI in Medicine 2025" — Writing % may be understated
Open
Thread #4421 · declare-ai.org
RK
The declaration lists 55% writing by Claude, but based on stylometric analysis using the linked tool, the estimated AI-written percentage is closer to 78–82%. I ran the article through three detectors independently. The human editing appears minimal. Happy to share my methodology. Requesting the declarer provide more detail on what sections were human-written.
JR
Fair challenge. I'll clarify: the 55% reflects the initial draft output from Claude. Sections 3, 5, and the entire Discussion were rewritten by me from scratch — these are not well captured by detectors because they draw on clinical experience. I'm happy to update the declaration to add a methodology note. Would 65% AI / 35% human with an annotation feel more accurate to the community?
CM
Marking this as Under Review. The declarer has engaged in good faith. If an updated declaration is filed within 72 hours, this thread will be marked Resolved. Standard procedure: community vote on updated declaration accuracy before badge is reset to green.
DISPUTE FILED VIA DECLARE AI API · THREAD AUTO-CREATED · 3 PARTICIPANTS · LAST ACTIVITY 8h AGO
🏛️
Forum Architecture
Self-hosted Discourse on a $6/mo VPS. Discourse SSO linked to declare-ai.org accounts so users don't need a separate login. Declare AI API creates threads programmatically via Discourse API when a dispute is filed. Categories: Disputes, Tool Updates, Governance, Feature Requests.
Discourse · Self-hosted
🗳️
Community Governance
Schema changes go through a 30-day RFC process in the forum. Any contributor can propose additions. Merges require 2 maintainer approvals. This mirrors the model used by W3C and CommonMark — slow by design, to preserve stability of the standard.
RFC Process

06 — Technology

Recommended stack

Chosen for: open source availability, developer familiarity, low ops overhead at launch, and ability to scale. Everything runs on commodity infrastructure a volunteer community can maintain.

LayerTechnologyWhy
Frontend Next.js 14 + Tailwind CSS SSR for SEO on declaration pages, React for dashboard UI, huge contributor base OSS
Widget Vanilla JS (zero deps, ~8kb) No framework overhead; embeddable on any site with a single script tag Performance
Extension WebExtensions API (MV3) Single codebase targets Chrome, Firefox, Edge, Safari with minor tweaks Cross-browser
API Node.js + Fastify Fast, low memory, excellent TypeScript support, familiar to most web devs Performance
Database PostgreSQL 16 JSONB for flexible schema storage, mature, excellent free hosting on Supabase/Neon Reliability
Cache Redis (Upstash) API response caching, rate limiting, session store. Serverless option avoids ops Low-ops
CDN + JSON Hosting Cloudflare R2 + Workers Free egress, global edge delivery of declare-ai.json files, KV for hot cache Free tier
Auth Auth.js (NextAuth) GitHub, Google, email magic link. No vendor lock-in, runs in the Next.js app OSS
Search Meilisearch Fast full-text search of declarations by URL, tool, contributor. Self-hostable OSS
Forum Discourse Industry standard for open-source communities. Ruby-based, rich API for automation OSS
CI/CD GitHub Actions Free for public repos, deeply integrated with GitHub where the code lives Free
Hosting (MVP) Fly.io or Railway Simple deployment from Dockerfile, free tiers adequate for launch, easy to migrate Simplicity
Monitoring Sentry + Grafana Cloud Error tracking and metrics. Both have generous free tiers for open source projects Free OSS tier

07 — Build Roadmap

Phase-by-phase to global standard

Each phase is independently shippable. The goal is to get something real in front of creators and developers as fast as possible, then grow through community contribution.

Phase 0 — Now · Complete
Concept, Schema Draft & Interactive Demo
Vision document (this page), full JSON schema v1.0 draft, interactive widget prototype in HTML/JS, architecture design. Ready to share with developers and post as public RFC on GitHub Discussions.
Schema v1.0 Draft Widget Prototype Arch Diagram Dev Briefing Doc
Phase 2 — Weeks 6–14
REST API, CDN Hosting & Browser Extension (MVP)
Backend API goes live. Declarations can be submitted and stored. declare-ai.json files hosted on Cloudflare CDN. Browser extension auto-detects declarations on any page and surfaces the widget. Submit to Chrome Web Store. Launch Discourse forum. Begin recruiting volunteer moderators.
REST API v1 PostgreSQL + Redis CDN Hosting Chrome Extension Discourse Forum
Phase 3 — Months 4–6
Dispute System, Accounts & Community Moderation
User accounts with GitHub/Google SSO. Dispute filing from the widget links directly to auto-created forum threads. Moderator tools and escalation paths. Community vote system for declaration accuracy. First cohort of verified organizations (universities, news orgs, open-source projects) onboarded.
User Accounts Dispute Flow Mod Tools Verification Badges Partner Orgs
Phase 4 — Months 6–10
Scoring Engine & Deepfake Detection
Debunk Score integrated from fact-check APIs. AI-text detection signals (aggregated, not single-model — too unreliable solo). Deepfake probability pipeline for linked video/audio. Firefox and Safari extensions. WordPress + Ghost + Substack plugins. Email notification system for disputes.
Debunk Score Deepfake API Firefox Ext WordPress Plugin Email Notifs
Phase 5 — Year 2+
W3C Proposal, Platform Integrations & Governance DAO
Submit declare-ai.json as a proposed W3C Note. Approach AI companies (Anthropic, Google, OpenAI, Mistral) to auto-embed declarations in exported content. Establish a formal open governance structure. Consider a protocol-level integration with the Internet Archive for permanent declaration storage.
W3C Note Platform Partners Governance Charter Archive Integration

08 — Open Source Strategy

Free forever, community sustained

The standard must be free to be universal. The infrastructure is funded through optional hosted services — nobody is ever forced to pay.

⚖️
MIT License
Widget, schema, validator, API, browser extension — all MIT. Anyone can fork, embed, redistribute without restriction. No copyleft complications for commercial sites that embed it.
🏗️
Self-Hostable
Full Docker Compose setup so organizations can run their own private Declare AI instance internally (hospitals, universities, governments) without sending data to our servers.
💰
Optional Revenue
Hosted dashboard for non-technical creators (~$5/mo) covers server costs. Verification API keys for high-volume platforms. Grants from Mozilla, Wikimedia, Open Society. No VC, no ads.
🤝
Contributor Model
GitHub Issues + Discussions for feature RFCs. All maintainers listed in MAINTAINERS.md. Good First Issue labels. Monthly community calls. Code of Conduct modeled on Contributor Covenant.
🌐
Localization
declare-ai.json schema and widget UI must be localized for non-English adoption. Crowdin for translation management. This is critical for global standard adoption.
📣
Launch Strategy
Phase 1 launch targets: Hacker News Show HN, Product Hunt, DEV.to post, outreach to 3–5 small publishers willing to pilot. Measure: declarations filed in month 1.
Repository to create: github.com/declare-ai/declare-ai — monorepo containing: /schema, /widget, /extension, /api, /site, /docs. Start with GitHub Discussions enabled and a pinned RFC thread for schema v1.0 feedback.

Declare AI
This Briefing Document
declare-ai.org/dev-brief-v1
AI
v1.0
AI Total
68%
Human
32%
Est. Hours
6.0 h
Tools
2
Writing
60%
Research
22%
Design
10%
Human
32%
Debunk Score
N/A
Deepfake
N/A
Transparency
Self-decl.
Disputes
0
Contributors
Vision Author Claude Sonnet