In March 2026, the TTB issued guidance on the use of AI-generated imagery in alcohol beverage advertising (read it here). The short version: AI-generated images used in alcohol marketing must comply with existing labeling and advertising regulations, including accurate representation of the product.

This sounds obvious. But there's a nuance worth understanding if you're choosing how to produce your product imagery this year.

The guidance applies broadly to any imagery categorized as AI-generated. The practical question it raises is how closely your vendor's process is anchored to your actual product.

There's a meaningful difference between two approaches.

A tool that generates a "wine bottle" from a text prompt or a rough photo produces something invented. A plausible bottle, not a representation of your real label, your specific closure, or your actual container. Whether it looks convincing is a separate question from whether it accurately represents your product.

A rendering process that starts from your label files, your exact closure type, and your actual bottle spec is producing a documented visual of your real product. The output is only as accurate as the files you provide — the same standard that has always applied to product photography.

Why does this matter for your marketing decisions?

If you're using generative tools to produce product imagery, your marketing team should be asking whether those outputs accurately represent what's in the bottle. If you're working with a rendering service that builds from your actual files and specs, you're in familiar regulatory territory. The accuracy standards haven't changed. The question is whether your process is built to meet them.

The practical upshot: not all "AI imagery" is the same thing legally, visually, or operationally. Know which kind your vendor is using.