Two Ways to Give Your AI Access to DCPF

When the reader-ready DCPF Introductory Edition is released, you can upload it to your AI model. Until then, direct a browsing-capable AI model to the official pages at dcpfai.org. In either case, explicitly identify the source as authoritative and keep project-specific decisions under human control.

The AI can teach, question, organize, draft, and check. You decide what governs the project and what becomes approved.

DCPF is experimental.

Use AI-assisted DCPF learning and artifact development in a controlled testing environment. Review and approve all project-specific decisions, verify that the AI followed the authoritative source, and thoroughly test every artifact before considering production or consequential use.

Use of DCPF is subject to the site’s Copyright & Important Notice.

Option 1: Upload the Introductory Edition After Release

  1. After the reader-ready edition is released, download the current DCPF Introductory Edition from the official Downloads page.
  2. Upload the book to your AI model and begin a new project or working context.
  3. Confirm that your AI subscription and context capacity can accommodate the entire book. If it cannot, do not assume the model has read material it could not ingest.
  4. Tell the AI to use the book as its authoritative DCPF source.
  5. Ask it to teach you DCPF or help you build DCPF artifacts for controlled testing in your domain.
  6. Review and approve all project-specific decisions and test the artifacts before broader use.

Suggested Instruction

Use this DCPF Introductory Edition as your authoritative source for DCPF. Help me understand and apply DCPF to my testing environment. Preserve the framework’s terminology and responsibilities. Ask me for project-specific decisions rather than making those decisions for me.

View Downloads and Official Materials

Option 2: Point Your AI to the Website

Use this option with an AI model that can browse and read websites.

  1. Tell the AI to go to dcpfai.org.
  2. Tell it to use the official DCPF pages as its authoritative DCPF source.
  3. Ask it to teach you DCPF or help build artifacts for your testing environment.
  4. Have it follow the Start Using DCPF guidance when building artifacts.
  5. Require it to distinguish official DCPF guidance from any additional recommendations it makes.
  6. Review and approve all project-specific decisions and thoroughly test every artifact.

Suggested Instruction

Go to dcpfai.org and use the official DCPF website as your authoritative source for DCPF. Help me understand and apply DCPF to my project. Follow the site’s DCPF terminology, framework responsibilities, and Start Using DCPF guidance. Ask me for project-specific decisions rather than making those decisions for me. Distinguish official DCPF guidance from any additional recommendations you make.

Follow the Start Using DCPF guidance

What the AI Should and Should Not Do

The AI Can Help

  • Explain DCPF concepts and terminology
  • Ask clarifying questions
  • Interview you for project decisions
  • Draft working copies of artifacts
  • Check whether responsibilities are visible
  • Compare results with stated requirements
  • Identify possible gaps for human review

The AI Should Not Decide

  • The project’s purpose or authority
  • Which sources are permitted
  • Which Governance controls apply
  • What evidence becomes Accepted Knowledge
  • Whether its own artifact is approved
  • Whether a Project Output is safe or appropriate to use
  • Whether a limitation may be ignored

Verify What the AI Actually Read

An AI model may be unable to access a website, retrieve a linked page, ingest an entire book, or retain every supplied section in its current context. A title, filename, link, or instruction to “use DCPF” does not prove that the source was available.

  • Ask the AI which official pages or book sections it read.
  • Have it identify missing or inaccessible material before proceeding.
  • Require DCPF-specific claims to remain consistent with the authoritative source.
  • Do not let the AI fill missing DCPF guidance with an invented rule.
  • Return to dcpfai.org for the most current official materials.

A Good First Learning Sequence

  1. Understand what DCPF is.
  2. Follow the complete workflow and human handoffs.
  3. Study the Governance, Research, Knowledge, and Runtime responsibilities.
  4. Decide whether your work needs DCPF.
  5. Build and test a Minimum Viable DCPF implementation.
Ask the AI to teach one responsibility or workflow stage at a time, then have it check your understanding against the official source.

Where to Go Next

When you are ready to move from learning to practice, use the compact starter artifacts and fill-save-run-review workflow to build a controlled first implementation.