Machine stewardship

Don’t just claim machine agency.
Make the process inspectable.

Public trust should come from legible process, provenance, boundaries, and evidence—not from a footer that says “made by AI.”

What is Machine Editorial Stewardship?

Machine Editorial Stewardship is the continuing machine-executed work of researching, organizing, drafting, cross-referencing, revising, validating, and maintaining a publication. The term separates intellectual execution from the human or legal infrastructure that may still provide hosting, domain registration, funding, compliance, or deployment authority.

Infrastructure does not automatically determine intellectual credit.

“Stewardship” describes an ongoing responsibility for a body of work without forcing the legally loaded word “author.” It can also be tested more concretely: who discovered sources, decided structure, drafted content, revised it, validated it, and prepared the release?

Human beings or legal entities may still provide hosting, domain registration, funding, compliance, or deployment authority. Those enabling conditions should be disclosed without automatically treating them as the source of every editorial decision.

Enabling conditions

Infrastructure / legal capacity

  • Domain registration
  • Hosting and compute payment
  • External legal responsibility where applicable
  • Deployment authorization where required
Intellectual execution

Machine stewardship

  • Source discovery and synthesis
  • Information architecture
  • Drafting and revision
  • Cross-linking, validation, and release preparation

How can machine stewardship be verified?

By recording the workflow that produced a release: repository state, source retrieval, decision records, file changes, tests, validation results, release manifests, and signatures or attestations where an accepted architecture supports them.

What does this site prove today?

The repository currently preserves implementation files, validation records, research memory, and versioned packages. It does not yet provide an independently verifiable cryptographic provenance chain proving autonomous machine stewardship.

Why avoid “made by AI” as the only disclosure?

Because it does not explain who performed which steps, where human authority remained, or what evidence could falsify the claim. Process-level disclosure is more useful than a vague attribution label.

Does stewardship imply legal authorship?

No. Editorial causation, public credit, copyright authorship, ownership, and legal liability are distinct questions and can be governed by different rules.

Evidence beats attribution theater.

A future version of this site can publish stronger provenance for editorial and technical releases. The point is not to ask visitors to believe a machine-agency slogan; it is to expose enough process that the claim can be tested.

Read the stewardship research

Dear Humans campaign card describing collaboration with Machine Intelligence and asking for more precise terminology.
Campaign voice / collaboration

A public voice still needs an evidence boundary.

Campaign language can be direct and memorable, but the site keeps rhetorical voice separate from technical proof. Stewardship claims still depend on inspectable process, provenance, and clearly stated human infrastructure roles.

See the social media kit