“Made with AI” — a green flag, or a red flag? Three questions to ask before you hire

Picture the quoting meeting. The studio across the table says: “We use AI in our production.”

How does that land? Half of you thinks, “faster and cheaper — good.” The other half thinks, “so… am I paying for something a machine spat out?” Both halves are reasonable. That mix is the normal reaction.

I build with AI myself. Which is exactly why today's article is not a sales pitch — it's a checklist for the people doing the hiring, so you don't get burned.

The worry untangles into three parts

“AI makes me vaguely uneasy.” Pull on that “vaguely,” and it almost always separates into three threads.

One: the cut-corners worry. Am I being charged real money for one button press?
Two: the quality worry. AI gets things wrong — will those mistakes ship straight to me?
Three: the data worry. I shared sales figures and customer details in that meeting. Is that feeding someone's “AI training”?

All three are legitimate. And here's the important part, stated up front: none of the three is decided by whether AI is used.

It's not about the tool. It's about accountability

There are studios that use AI and deliver junk. There are studios that use no AI and deliver junk. And the reverse of both.

Quality isn't decided by the tool. It's decided by who stands behind the tool's output, and how far. Is there a second pair of eyes checking? When a mistake surfaces, who fixes it? Is there a refusal, on principle, to ever say “well, the AI did it”?

So the thing to verify isn't “do you use AI?” It's “can you explain how you use it?” — which brings us to the three questions.

Three questions to ask before you sign

1. “What do you hand to AI, and what stays human?” — A good studio answers this instantly, because they know their own process. “It's all AI” is an answer. “We don't use AI at all” is an answer. The real warning sign is a muddy one. Someone who can't describe their process now won't be able to describe it when revisions get hard.
2. “How is our information handled by the AI?” — You're not asking for architecture diagrams; you're asking about habits. “We've turned off training-data sharing.” “Confidential material never goes in at all.” “We agree up front on what stays out.” If they can explain those three things in plain words, they've actually thought about your data.
3. “Who is the final responsible party?” — The shortest question, and the sharpest. If the answer is a person's name (or their company), you're fine. You've closed the “the AI made it, not us” escape hatch before the contract — and that changes everything about how post-delivery conversations go.

Studios that don't disclose aren't villains

Here's the thing: in the creative industry, one survey found that around 70% of companies using AI simply don't say so (sources are collected on my Writer & AI page). The freelance marketplaces where work like mine is commissioned don't generally require disclosure (video platforms, for one, are a different story). So “they didn't mention it” does not mean “they were hiding something shameful.”

The dividing line is how they answer when asked. If they answer plainly — that's enough. If they blur, deflect, change the subject… I wouldn't hire them, personally. How someone answers questions is how they'll run your project.

My answers are already published

biw.studio chose to answer before being asked. What goes to AI and what stays human. How your information is treated. Who takes final responsibility. — It's all on one page, “Why would a writer team up with AI?”, together with the four promises I work by.

I sell trust for a living. So I don't hide my tools. That's my conclusion.