BRENNEN

Industries / Retail, hospitality and personal services / Beauty salons and spas / Tattoo studios

Design

Design is the work of drawing the image, fitting it to a body and putting its outline on the skin as a guide before the needle starts.

6

Milestones achieved, from c. 1900 to 2024.

1

To watch for: announced, not yet confirmed outside.

2

Open deliverables. Nobody has shipped them.

The road

AchievedWatch forOpen

Achieved

  1. c. 1900

    Plastic stencils carry designs

    A stencil puts a design's outline on the skin as a guide for the needle. By about 1900, by the Tattoo Archive's account, plastic stencils were in almost every tattoo shop in the industrialized world: the outline was scratched into a sheet of celluloid or acetate with a needle held in a pin vice, powdered charcoal was rubbed into the scratch, and the sheet was pressed onto the skin. A pencil rubbing taken from a stencil could be mailed to another tattooist, which is how designs travelled between shops before magazines, conventions and the internet.

    LessonSome shops were sold cellulose nitrate, which burns, and a 1940s Portland, Oregon newspaper blamed a fire in George Fosdick's shop on his stencils.

    tattooarchive.com

    Credit

    tattooists of the period, as recorded by the Tattoo Archive in Winston-Salem, North Carolina

  2. 1980s

    Stencils copied by heat

    A thermal copier, which tattooists call a thermofax, prints a line drawing onto transfer paper in seconds, so a stencil no longer has to be cut by hand. The Tattoo Archive says these machines changed how tattooists worldwide transferred designs in the 1980s. By the time the archive wrote its account, some suppliers put a thermal copier in beginners' kits rather than an autoclave, the steam sterilizer a studio needs, and many tattooists had never worked with a plastic stencil.

    tattooarchive.com

    Credit

    tattooists who adopted office copiers for stencils, as recorded by the Tattoo Archive

  3. 2011

    A face tattoo in court

    S. Victor Whitmill designed and applied the tattoo on the boxer Mike Tyson's face in Las Vegas in 2003. He says Tyson signed a release giving him ownership of the design, which he registered for copyright. In April 2011 he sued Warner Bros. in federal court in St. Louis for copying it onto the actor Ed Helms's face in The Hangover Part II, and asked the court to stop the film. The judge let the film open as scheduled on 26 May 2011, and Warner Bros. said on 20 June 2011 that the two sides had settled, on terms not disclosed.

    LessonThe case made tattoo copyright news but ended without a ruling on whether a tattoo can be copied without the tattooist's permission.

    abajournal.comcbsnews.com

    Credit

    S. Victor Whitmill; Warner Bros

  4. 2020

    Game tattoos ruled fair use

    Solid Oak Sketches held the copyrights in five tattoo designs on three NBA players and sued 2K Games and Take-Two Interactive for showing them in the NBA 2K basketball games. On 26 March 2020 the US District Court for the Southern District of New York granted the publishers summary judgment, a ruling without a trial, finding the use was fair use, the defense in US law for some uses made without permission. By the court's figures the tattoos appeared on 3 of about 400 players and made up 0.000286% to 0.000431% of the game data, and no evidence showed a market for licensing tattoos to games.

    copyright.gov

    Credit

    US District Court for the Southern District of New York

  5. 2022

    Infringed, but no damages

    Catherine Alexander tattooed the wrestler Randy Orton between 2003 and 2008 and sued WWE and Take-Two in 2018 after her designs appeared on his character in the WWE 2K games; according to her suit, WWE had offered her $450 for the rights and she refused. On 30 September 2022 a federal jury in the Southern District of Illinois found the copying was not fair use and awarded her $3,750. In September 2024 Judge Staci M. Yandle let the finding stand, noting a game feature that let players put her designs on their own characters, but struck the damages because no evidence at trial supported them.

    LessonA tattooist can win on copyright and still recover nothing without proof of what the copying cost.

    f4wonline.comcopyright.govblog.ericgoldman.org

    Credit

    Catherine Alexander; Judge Staci M. Yandle

  6. 2024

    Jury clears a traced portrait

    In 2017 the tattooist Kat Von D (Katherine Von Drachenberg) traced Jeffrey Sedlik's 1989 photograph of Miles Davis to make a stencil for a client's tattoo, then shaded it freehand. At trial in January 2024 the jury found the tattoo, her sketch and most of her social media posts not substantially similar to the photograph, and found that four posts she agreed were similar, showing the photograph during the work, were fair use. The Ninth Circuit Court of Appeals upheld the verdict on 2 January 2026; that ruling was vacated in June 2026 for a rehearing (see Watch for).

    cll.commorganlewis.com

    Credit

    Kat Von D, Kat Von D, Inc. and High Voltage Tattoo; Jeffrey Sedlik

Watch for

  1. Full court rehears traced tattoo

    Sedlik argues that the jury's verdict on the tattoo traced from his photograph should be overturned. The panel that upheld it in January 2026 relied on the Ninth Circuit's "intrinsic test", in which a jury judges whether two works share the same "total concept and feel", and two of its judges wrote that the test should be reconsidered. On 9 June 2026 a majority of the circuit's active judges ordered the case reheard en banc, by a larger panel of eleven judges, and vacated the panel's opinion. Argument was set for the week of 28 September 2026 in San Francisco.

    Achieved whenthe en banc court issues its decision on the verdict.

    caselaw.findlaw.commorganlewis.commondaq.com

    Claimed by

    Jeffrey Sedlik, appellant; rehearing ordered by the US Court of Appeals for the Ninth Circuit, June 2026

Open

  1. 01

    Tattooists judge AI designs

    A public set of at least 1,000 designs, half made by AI image generators from everyday customer requests and half drawn by working tattooists, each judged by at least three tattooists who see the design at a stated size on the body and answer one fixed question: can it be tattooed as drawn, with small changes, or not at all, and why. The judges look at images on a screen; nobody is tattooed.

    The brief
    Credit · unclaimed
  2. 02

    Healed spread, measured

    A measured table of how much tattoo lines widen as they heal and age, by needle grouping, ink color and body area, built from records studios already hold: the stencil file, a photo taken on the day and a photo taken at a later visit. From the table, a rule designers can use, such as the smallest gap between two lines that stays open after five years at a given size. Only existing photos are used, each with the client's written consent and an ethics review; nobody is tattooed or examined.

    The brief
    Credit · unclaimed

Open 01 · the brief

Tattooists judge AI designs

Credit · unclaimed

The first to publish is credited on this line, and the deliverable moves from Open to Achieved.

What to build

A public set of at least 1,000 designs, half made by AI image generators from everyday customer requests and half drawn by working tattooists, each judged by at least three tattooists who see the design at a stated size on the body and answer one fixed question: can it be tattooed as drawn, with small changes, or not at all, and why. The judges look at images on a screen; nobody is tattooed.

Why it is still open

AI tattoo generators are usually ranked with reward models, which predict which image a crowd of annotators prefers, while a worn tattoo also has to survive a needle, gaps that heal shut and ink that spreads. In October 2026 Yuzhu Cong, who runs an AI tattoo generator, published an audit of four reward models and of physical "probes" (thinnest line width, detail finer than a 0.30 mm needle, gaps under 0.50 mm) on 1,348 tattoo-prompt images from a public preference dataset. The probes agreed with crowd preference below chance, and the only human labels in the study were crowd preferences, not tattooists' judgments.

zenodo.orghuggingface.co

Built with what exists

Open image models and the tattoo-style adapters shared on Hugging Face; Cong's open probe code and indexes (github.com/Stark-Will/tattoo-probes, CC-BY-4.0 data); Label Studio, an open-source labeling tool, or a plain web form for the judging; Python for agreement statistics.

  1. First step

    Write 50 customer-style requests, generate two designs for each with an open model, and have three tattooists judge the first 100.

  2. Done looks like

    1,000 designs, each with three tattooists' verdicts and reasons; agreement between judges reported as Fleiss' kappa, a standard measure of how far several raters agree; a table of how well each reward model and each probe predicts the tattooists' verdict; the set released under an open licence.

Open 02 · the brief

Healed spread, measured

Credit · unclaimed

The first to publish is credited on this line, and the deliverable moves from Open to Achieved.

What to build

A measured table of how much tattoo lines widen as they heal and age, by needle grouping, ink color and body area, built from records studios already hold: the stencil file, a photo taken on the day and a photo taken at a later visit. From the table, a rule designers can use, such as the smallest gap between two lines that stays open after five years at a given size. Only existing photos are used, each with the client's written consent and an ethics review; nobody is tattooed or examined.

Why it is still open

The 2026 tattooability audit modelled spreading ink as an even blur of 0.30 mm and states that this model is "not validated on healed skin". The best-known earlier work, Ian Eames's 2011 model at University College London, predicted that fine detail can lose its definition over 15 years; it was a model, not a set of measurements.

huggingface.colivescience.com

Built with what exists

Archived studio photos and stencil files; the stencil's known size as the scale; OpenCV or Fiji (ImageJ) to align each stencil with its photos and measure line width; Python for the growth fit.

  1. First step

    Ask three studios for 20 archived cases each where the client has agreed in writing to research use, and measure the first ten by hand.

  2. Done looks like

    At least 300 tattoos with a stencil, a day-one photo and a healed photo at least one year later; line width measured at 20 points on each; a fitted widening rate in millimetres per year with confidence intervals by ink color and needle grouping; a second person repeating 30 tattoos within 0.05 mm.