Skip to content

AI production records for creative directors

You do not run the generations yourself, but you sign the work, and everyone in the room treats your signature as the answer to how it was made.

Published July 22, 2026.

A creative director's exposure is simple: approval implies knowledge. When a client asks which parts of a campaign are AI-assisted, an answer like 'some of it' does not survive a procurement review, a nervous brand team, or a journalist's follow-up question. The person who approved the work is expected to know what the work is, and in 2026 that includes how it was produced.

The problem is visibility. Work reaches a creative director as comps, boards, and finals, while the generative steps happened earlier, in other people's tools and accounts. Without a record attached to the asset, the creative director's knowledge of the process is secondhand at exactly the moment they are asked to vouch for it firsthand.

Records also change the review itself, which is the underrated part. Knowing whether a new version was hand-edited or regenerated from scratch tells you whether your feedback actually landed or was worked around. Knowing which references went into a board matters before you present it, not after, especially when a client's competitor could plausibly be sitting in the reference set.

It is equally important to know what a record cannot do. It can show what existed, that it has not changed since, and when it entered the record. It cannot show that a particular reference caused a particular output, so never promise a client that kind of causal certainty. The provenance guide linked below draws that line clearly, and it is worth internalizing before a client meeting tests you on it.

Where documentation breaks down

  • Approvals happen on the finished asset, but the questions that follow are about the process behind it, which the creative director never saw.
  • Reference boards get assembled in a hurry during pitches, and nobody logs what went into them, which turns awkward the moment a client asks.
  • Feedback rounds blur the line between editing and regenerating, so nobody can say later which version was reworked and which was quietly replaced.
  • Team members reach for tools that were never approved for the project, and the creative director learns about it when the client asks about a product they have never heard of.
  • The hardest questions come in live settings, a client meeting or a panel, where there is no time to go and check.

A documentation routine that holds up

  1. Set an approved-tools list at project start and put it in the brief, so the answer to 'what did we use' is decided before anyone generates.
  2. Make the record a condition of review: a draft arrives with its generation record attached, or it waits until it does.
  3. When giving feedback, say whether you expect an edit or a regeneration, and check the record afterwards to see which one you got.
  4. Before any client presentation, read the per-asset summaries, so what you say in the room matches what the record says on paper.
  5. At sign-off, approve the record together with the asset, so approval and evidence become the same event instead of two that drift apart.

Frequently asked questions

A client asks in a meeting which parts of the campaign are AI-generated. What lets me answer on the spot?

A per-asset summary that was maintained during production, read before the meeting. Improvised percentage estimates feel helpful in the room, but they become commitments the record has to bear later, so only say what the record supports.

My team says record-keeping slows down exploration. Are they wrong?

They are right about exploration and wrong about production. Let exploration stay loose, and draw the line at exposure: the moment an output is shown to a client or enters a deliverable, it needs its record. That single rule keeps the burden where the risk is.

Can records tell me whether my feedback was actually applied?

In one specific and useful way, yes. A record shows whether the next version was an edit of the previous one or a fresh generation, and those are two different responses to the same note. It will not judge the creative quality of the response, but it ends the ambiguity about what happened.

Do I need to disclose AI use in pitch work?

That depends on the contract and on the rules that apply to the client's market, so check both rather than assuming. Keep the record either way, because pitch assets regularly graduate into production, and the questions arrive when reconstruction is no longer possible.

Who owns the record, me or the producer?

The producer runs it day to day, but the creative director is the person most damaged when it is missing, because the questions come to whoever signed. Your part of the ownership is making the record a condition of review, and that costs you one sentence per project.

Related guides

  • How to document AI-generated work for client delivery What clients and platforms already ask about delivered work, which is what they will ask you.
  • AI provenance: what you can prove and what you cannot The line between an answer you can stand behind and one that unravels, drawn before the meeting.
  • EU AI Act Article 50 for creative teams The transparency obligations in plain terms, for the person who ultimately signs the work.
  • Prompt and model documentation: a practical checklist for studios The fields your team should be capturing, so you know what to demand at review.
  • What is an AI production log? An AI production log is the per-asset record of prompts, models, settings, references, edits, rights decisions, and approvals behind AI-assisted work. Here is the minimum useful structure and the limits of what that record can prove.
  • C2PA vs. AI workflow documentation C2PA and AI workflow documentation solve different parts of the provenance problem. C2PA protects signed assertions attached to media; workflow records preserve prompts, sources, decisions, rights, versions, and approvals around the file.
  • AI shot log template for image and video production A practical AI shot log template for prompts, models, references, versions, human edits, disclosure, and approval. Download the CSV and adapt the field guide to your production workflow.
  • How to record client approval for AI-generated work A client approval record for AI-generated work should identify the exact version, disclosed AI use, review scope, requested changes, approver, and time. This guide turns sign-off into evidence instead of an ambiguous email.
  • Guide editorial and verification policy How Behind The Workflow researches, sources, dates, updates, and corrects its AI production guides. Product facts use primary sources, legal claims link to official text, and machine-readable markup mirrors visible content.
  • AI production records for AI producers AI producers own the delivery questions nobody else can answer: which model, which prompt, which version, approved by whom. This guide sets out the records that let a producer answer from evidence instead of memory.
  • AI production records for post-production teams In post, AI arrives as a hundred small fixes inside shots rather than one big generation, which is exactly what makes it hard to declare on a delivery form later. This guide covers records that map to shots, versions, and conforms.
  • AI production records for freelance AI artists Freelance AI artists warrant their own process in every contract they sign, with no legal department behind them. This guide covers the records that back those warranties, survive tool subscriptions, and win repeat work.
  • AI production records for agencies Agencies carry contractual responsibility for AI use across every team, freelancer, and production partner on an account. This guide covers records that hold at agency scale, where the person who generated and the person who answers are never the same.

Every requirement in this guide is answered from a record kept while the work happens. BTW keeps that record for you.

Start your own record