What to look for in a creative partner when AI is part of their workflow

(This article is for informational purposes only and does not constitute legal advice. Consult a qualified attorney for guidance on IP ownership and AI-generated content.)

The question used to be “do you use AI?” That question no longer tells you much. According to Envato’s 2026 State of AI in Creative Work report (n=1,780), 49% of creative professionals now use AI daily for client work. The tool is everywhere. What actually protects your project is a different set of questions: how are they using it, at what stage, and on which parts of your deliverable?

Getting this right matters more than it might seem. When clients discover undisclosed AI use on their own, research suggests the trust damage is significantly worse than if the partner had disclosed it proactively (Schilke & Reimann, 2025, Organizational Behavior and Human Decision Processes). And 82% of global brands now say AI transparency is essential to how they select creative partners (World Federation of Advertisers, April 2026).

The good news: you can evaluate this before you hire.

The disclosure gap is real

Most creative professionals are using AI. Far fewer are telling their clients about it.

The Envato data is specific: 58% of creative professionals have used AI for client work without disclosing it. Only 28% of agency owners say they always inform clients. That gap between use and disclosure is not a minor oversight. It affects what you own, what you can verify, and whether the work you paid for reflects the expertise you thought you were hiring.

This is not a reason to avoid working with anyone who uses AI. It is a reason to ask about it directly, before the project starts.

Why it matters for what you own

Here is something that affects every client commissioning creative work in 2026: solely AI-generated works cannot be copyrighted under U.S. law.

In March 2026, the U.S. Supreme Court declined to hear Thaler v. Perlmutter, affirming the existing standard set by the U.S. Copyright Office. Works with meaningful human authorship embedded in them can still qualify for copyright protection. Works where a human prompted an AI and approved the output, without meaningful creative contribution, likely cannot.

This means the final deliverable your creative partner hands you may or may not be ownable, depending entirely on how it was made. Asking your partner how they document human authorship is not a technical question. It is a business one.

(This article is for informational purposes only and does not constitute legal advice. Consult a qualified attorney for guidance on IP ownership and AI-generated content in your specific context.)

The three questions that actually tell you something

Forget the binary. The evaluation that protects you has three parts.

How: Is AI assisting a skilled human, or is it generating the output? There is a meaningful difference between a designer using AI to iterate on layout options they then refine, and a designer who prompts an image generator and hands you the result. The red flag is not AI use. It is minimal human creative input at the decision-making stage. If the partner cannot explain what they personally contributed and why, that is worth noting.

When: At what stage does AI enter the process? A partner who uses AI to generate initial concepts for a brainstorm presents a different risk profile than one who uses it to produce the final visual or the published copy. AI at the ideation stage, feeding into human judgment and craft, is a very different arrangement than AI producing the final deliverable. Ask specifically: where in your process does the human hand take over?

On what: Which parts of the deliverable are AI-touched? Draft copy, final images, concept strategy, and brand voice guidelines each carry different implications. A partner who uses AI to draft a first pass at body copy and then rewrites it substantially is not the same as one who uses it to generate your brand positioning. Get specific about which elements are AI-assisted and which are not.

Questions to ask before you hire

These are designed to be copied into your next discovery call or creative brief. (See also: what to ask before hiring an AI-assisted creator)

1. “Which AI tools do you use, and for which parts of this project?”

A confident, specific answer tells you the partner has thought about this deliberately. Vague answers (“we use AI where it makes sense”) should prompt a follow-up.

2. “Do you disclose AI use in your project documentation or invoicing?”

This tells you whether transparency is built into their process or whether it only happens when asked. Proactive disclosure is the higher standard.

3. “How do you ensure meaningful human authorship in the final deliverable?”

This is the IP question in plain language. A partner who has a clear answer, whether that is a documented review process, a policy on which outputs are AI-assisted, or a commitment to human-led concept development, understands what is at stake.

4. “What is your process when the brief requires original, ownable creative concepts?”

This surfaces whether they treat AI-assisted work and original creative work as distinct categories. Good partners do. They reserve human-led originality for work that needs to be ownable.

5. “If I ask to see how a piece was created, what can you show me?”

Process transparency should not be a surprise request. A partner who can walk you through their workflow, including where AI was used and where it was not, gives you meaningful accountability.

Green flags and red flags

Green flag: Your partner can explain their AI workflow unprompted, including which tools they use and for what.

Red flag: They are vague about tools or give a non-answer like “we use AI where it adds value” without specifics.

Green flag: AI handles early-stage tasks like research, rough drafts, or iteration; the human handles strategy, judgment, and final craft.

Red flag: AI generates the final output and the human role is review and approval, with minimal creative input in between.

Green flag: Your partner has a documented approach to IP and can tell you clearly what in the deliverable is ownable.

Red flag: They have no clear answer on copyright implications, or they dismiss the question with “it shouldn’t matter for your use case.”

What this looks like in practice

At The Blue Mango, every creator in the co-op goes through an evaluation before they work with clients. Part of that process covers AI: what tools they use, at which stages, and how they document human authorship in their work. We also ask where human judgment is non-negotiable for them, and what they will not hand off to an AI tool regardless of efficiency.

This is what precision hiring for creative services looks like in practice: criteria vetted upfront, not discovered mid-project. When you match with a TBM creator, you are not starting from zero on this conversation. The vetting has already happened. You can still ask your own questions (and you should), but you are not evaluating a partner’s AI transparency for the first time in a sales call.

The IAB released the industry’s first voluntary AI Transparency and Disclosure Framework in January 2026. Matching platforms and agencies that take this seriously are starting to build these standards into their intake processes, not leave them to clients to discover mid-project.

For clients commissioning creative work this year, the shift worth making is from “do they use AI?” to “have they thought carefully about how?” The answer to the first question is almost always yes. The answer to the second one varies considerably.

If you want a creative partner whose AI workflow has already been evaluated, talk to us. We can match you with a creator whose process aligns with what you need to own and publish with confidence.

(This article is for informational purposes only and does not constitute legal advice. Consult a qualified attorney for guidance on IP ownership and AI-generated content.)