# The Manifesto for AI-Driven Development

Four values and twelve principles for developers who build software with AI as a deliberate partner.

## Preamble

As AI-Driven Developers, we are discovering better ways of building software by working with AI as a deliberate partner, and helping others do the same.

# Values

## Value 1: Method over Model

Bet on the method, not the model — every LLM release will fade. Your method endures and standardizes your team’s practices across models, vendors, and versions.

## Value 2: Ownership over Delegation

You own what you ship — even what the AI wrote. Every commit is signed by a human; the AI accelerates, you decide.

## Value 3: Understanding over Acceptance

Don’t accept what you don’t understand. The AI is your collaborator — not your replacement.

## Value 4: Outcome over Output

Writing code is easy. Building product is not. Ship small, iterate fast, keep feedback loops short — avoid vanity metrics. The only number that serves is the business.

# Principles

## Principle 01

We deliver working software to production, faster and without debt.

Speed and quality are not in conflict. They are the same discipline, practiced twice.

## Principle 02

Planning is our first act of collaboration with AI.

The quality of our plan determines the quality of our output. Everything downstream inherits it.

## Principle 03

We break complexity down until it can be delegated cleanly.

Clarity of scope is clarity of output. Ambiguity compounds; decomposition dissolves it.

## Principle 04

Context is our primary asset.

We invest in it, maintain it, and protect it from noise. A polluted context produces polluted code.

## Principle 05

We apply the same standards to AI-generated code as to any code we write by hand.

There is no lower bar for generated work. Review, test, reject — the same rituals apply.

## Principle 06

We are the architects. Every line that ships is our line — regardless of what produced it.

Authorship is not about the keystroke. It is about the judgment that sent the keystroke.

## Principle 07

We use AI to go further in what we already understand.

We do not delegate what we have not mastered. Competence first; acceleration second.

## Principle 08

When a session loses coherence, we restart.

A clean context is always cheaper than a compacted one. Fresh starts are not failures — they are hygiene.

## Principle 09

We choose a consistent methodology over a collection of tools.

The method outlasts any model. Tools rotate; practice accrues.

## Principle 10

What one of us discovers, all of us adopt.

The community is our compound learning machine. Hoarded technique decays; shared technique compounds.

## Principle 11

We recognize AI's limits as signals — not walls to push through, but invitations to decompose further.

When the model stalls, the prompt is wrong. When the prompt is wrong, the scope is wrong.

## Principle 12

We do not stop learning because AI can code. We learn faster because it can.

The leverage is real. What it multiplies is up to us.

## Sign

Signing is a public act recorded in the source repository. One YAML file, one pull request, no backend.

Repository: https://github.com/ai-driven-dev/manifest
Canonical URL: https://www.ai-driven-development.org/