---
url: /design-history/graphic-design-in-the-age-of-ai/
title: "Graphic design in the age of AI | Design History | TGDS"
template: design-history-movement
priority: 3
wordCount: 2011
lastModified: 2026-06-22T06:00:35.134Z
category: pages
site: "The Graphic Design School"
tokenCount: 2802
---

# Graphic design in the age of AI | Design History | TGDS

Design history · Contemporary Graphic design in the age of AI The generative-AI shift of the early 2020s — what changed for designers, and the questions still open. Between April and November 2022, generative AI moved from research demos into working designers' hands: DALL·E 2, Midjourney, Stable Diffusion and ChatGPT all reached the public inside eight months. The shift was real and datable. What it means for graphic design is not settled. This page traces the tool releases, the moment the design world first argued about them, the copyright and authorship questions still working through the courts and the US Copyright Office, and the industry's first licensed-data response — and keeps the open questions framed as open. Key facts Inflection Generative AI reaches working designers — Apr–Nov 2022 Tools DALL·E 2 (Apr 2022) · Midjourney open beta (Jul 2022) · Stable Diffusion (Aug 2022) · ChatGPT (Nov 2022) Flashpoint Jason Allen's *Théâtre D'opéra Spatial* (Midjourney) wins the non-professional division of the Digital Arts / Digitally-Manipulated Photography category at the Colorado State Fair, 29 Aug 2022 Open questions Training-data copyright · authorship & registrability · attribution · licensed-data tooling Industry response Adobe Firefly (Mar 2023) ships generative AI inside the standard toolchain on a "commercially safe" / licensed-data positioning Status Unsettled — a live shift in practice, not a closed chapter 01 History & context For most of graphic design’s history, a new tool arrived slowly enough to be absorbed. Movable type spread over decades, photo-typesetting over years, desktop publishing over a single decade. Generative AI arrived over eight months. Between April and November 2022, four tools reached the public. DALL·E 2 opened in beta from OpenAI. Midjourney moved to open beta on 12 July, running inside Discord. Stable Diffusion was released as an open model on 22 August, with its weights public so anyone could run it locally. ChatGPT followed on 30 November. By the end of the year, a designer could type a sentence and get an image, or a brief, or a hundred name options, with no intermediate craft step. The design world’s argument crystallised before ChatGPT even shipped. On 29 August 2022, Jason Allen’s Théâtre D’opéra Spatial, made in Midjourney, edited in Photoshop and upscaled with Gigapixel AI, won first place in the emerging-artist digital-art category at the Colorado State Fair. It was among the first generative-AI images to win such a prize, and the objection from working artists was immediate. The Fair began requiring entrants to disclose AI use the next year. This page sits in the contemporary cluster because that is what it is: a live shift in current practice, not a closed historical chapter. The facts below are datable and sourced. The questions they raise are not yet answered. 02 What actually changed The shift is easier to describe by what it disrupted than by any shared style. The craft step became optional. Between brief and image there used to be drawing, photography, typesetting, retouching: skill that took time and could be charged for. A prompt collapses that step. The question this raises for a designer is not “can the tool make an image” (it can) but “what is the designer’s contribution now, and is it defensible.” The training data is contested. The open image models were trained on billions of images scraped from the web, including copyrighted work, without permission or payment. Whether that training is lawful is the subject of active litigation (below). A designer using these tools is working on top of an unresolved legal question. Authorship is unsettled. A purely AI-generated image cannot, on current US guidance, be copyrighted — there is no human author. That changes what a studio can own and license. It is a practical constraint, not a philosophical one. The toolchain absorbed it. Within a year, generative AI was not a separate app but a feature inside Photoshop — Generative Fill shipped as a standard tool in September 2023. The decision to use it stopped being a deliberate choice and became a default in the box. 03 The legal and authorship questions These are the questions the field is still working through. They are framed here as open, because they are. Training-data copyright. On 13 January 2023, artists Sarah Andersen, Kelly McKernan and Karla Ortiz filed a class action, Andersen v. Stability AI, against Stability AI, Midjourney and DeviantArt. It centred on the LAION image set used to train the models. In February 2023, Getty Images sued Stability AI in the US (with a parallel UK action) over the scraping of millions of its licensed images. Both tested, rather than settled, how copyright applies to model training. The Andersen case was still in active litigation, with no final ruling on the merits, well into 2026. The UK strand reached judgment first: on 4 November 2025 the High Court rejected Getty’s copyright claims, holding that a model’s trained weights are not “infringing copies” — an early, jurisdiction-specific result rather than a general answer. Authorship and registration. On 21 February 2023, the US Copyright Office ruled on Zarya of the Dawn, a graphic novel by Kris Kashtanova: the Midjourney-generated images were held not protectable, while Kashtanova’s text and the selection and arrangement of elements were. On 16 March 2023, the Office issued formal registration guidance — copyright protects only human-authored material, purely AI-generated output is not registrable, and more-than-de-minimis AI content must be disclosed on an application. In March 2025, the US Court of Appeals for the DC Circuit affirmed that principle in Thaler v. Perlmutter: a work authored entirely by a machine, with no human author, cannot be registered — the first appellate decision to settle the point in US law. A licensed-data response. In March 2023, Adobe announced Firefly, built into Photoshop and Express. Its first model was trained on Adobe Stock, openly-licensed content and public-domain material, and was positioned as “commercially safe,” a direct answer to the training-data questions hanging over the open models. The positioning was not unchallenged: reporting in April 2024 said a small share of Firefly’s training images were themselves AI-generated (including Midjourney output), which complicates a clean “licensed-data only” claim. Whether “trained on licensed data” fully resolves the authorship and ownership questions is itself still debated. Key works & examples DALL·E 2 (OpenAI) 2022 OpenAI announced DALL·E 2 in April 2022 and opened broad beta access in July 2022 — a text-to-image model that put usable generative imagery in front of a general audience for the first time. It set the template the rest of the year followed: type a prompt, get an image. Midjourney open beta 2022 Midjourney moved to open beta on 12 July 2022, running through Discord. Its distinctive, painterly default look made it the tool behind much of the year's most-shared AI imagery — and the tool used to make the work that triggered the design world's first real argument about AI. Stable Diffusion public release 2022 Released as an open model on 22 August 2022 by CompVis (LMU Munich) and Stability AI. Because the weights were public, anyone could run it locally and build on it — which is why it sits at the centre of the training-data copyright suits that followed. ChatGPT 2022 Released on 30 November 2022. A text tool rather than an image tool, but it pulled generative AI fully into the mainstream and into design workflows — copywriting, briefs, naming, ideation — within weeks. Cosmopolitan "AI" cover (Karen X. Cheng) 2022 For its June 2022 issue, Cosmopolitan ran what it billed as the first magazine cover designed by AI — made with DALL·E 2 by creator Karen X. Cheng working with the magazine's editors. The prompt was documented and published alongside the cover. It is an early, datable example of a generative tool producing a finished, commercial graphic-design artefact rather than a one-off image. The Economist "AI's New Frontier" cover 2022 The Economist's June 2022 cover, made with Midjourney, was among the first uses of a generative tool on a major news title. It was later acquired by the Victoria and Albert Museum (collection record O1763062) as an early artefact of generative AI — an unusually clear marker that the design world treated these covers as historically significant. Théâtre D'opéra Spatial (Jason M. Allen) 2022 Made with Midjourney, edited in Photoshop and upscaled with Gigapixel AI; Jason Allen's piece won first place in the non-professional division of the Digital Arts / Digitally-Manipulated Photography category at the 2022 Colorado State Fair on 29 August — among the first generative-AI images to take such a prize. The win drew wide objection from artists and became the design world's flashpoint for the AI debate. The Fair began requiring AI disclosure from entrants the following year. Adobe Firefly 2023 Announced in March 2023, Firefly was Adobe's generative-AI family, built into Photoshop and Express. Its first model was trained on Adobe Stock images, openly-licensed content and public-domain material — positioned as "commercially safe," a direct response to the training-data questions raised against the open models. It put generative AI inside the standard designer's toolchain rather than a separate app. Firefly reached general availability on 13 September 2023, when Adobe also offered enterprise customers IP indemnification for Firefly-generated output — putting its "commercially safe" claim on a contractual footing. 04 Where this leaves the designer It is too early to write the legacy of a shift this is still happening inside. What can be said honestly is narrower. The tools are now part of the standard toolchain, and a working designer will encounter them whether or not they seek them out. The legal and authorship questions are live, not closed — a studio that builds a deliverable on purely AI-generated output is building on unsettled ground, and the prudent response is to know that rather than assume it away. The skills that survive the shift are the ones the prompt does not replace: judgement about what a piece is for, the ability to read a brief and a client, typographic and compositional discrimination, and the editorial sense to know when an output is right and when it is merely plausible. A design education teaches exactly those things, and they become more load-bearing, not less, when the production step gets cheap. Whether AI ultimately deskills the field or raises the floor under it is a genuine open question, argued in both directions by serious people. This page does not take a side on it. It records what happened, when, and from sources you can check. Learn at TGDS We teach the judgement the tools do not replace — typography, layout, brief-reading and editorial discrimination — and treat AI as a tool to be understood rather than feared or worshipped: Courses Certificate IV in Design (CUA40725) — our flagship qualification. The craft and judgement modules are exactly the skills that remain load-bearing when production gets cheap. Intensive Foundation Course — the 11-module foundation: typography, layout and composition fundamentals. Certificate of completion, not the Cert IV. Related movements & people Stefan Sagmeister Paula Scher Marian Bantjes Michael Bierut Sources & further reading Every claim on this page traces to one of these. Dates and outcomes are from primary or authoritative sources; nothing here asserts how the open legal questions will resolve. Online Smithsonian Magazine — AI art wins at the Colorado State Fair Washington Post — He used Midjourney to win a fine-arts prize. Did he cheat? US Copyright Office — Zarya of the Dawn letter (21 Feb 2023, PDF) Federal Register — Copyright Registration Guidance: Works Containing Material Generated by AI (16 Mar 2023) Copyright Alliance — Takeaways from Andersen v. Stability AI Mesh IP Law — Getty Images v. Stability AI litigation tracker Adobe — Adobe unveils Firefly, a family of new creative generative AI Stability AI — Celebrating one year of Stable Diffusion Cosmopolitan — The world’s first artificially intelligent magazine cover Victoria and Albert Museum — The Economist “AI’s New Frontier” cover (collection record) UK Judiciary — Getty Images v. Stability AI judgment (4 Nov 2025) US Copyright Office — Thaler v. Perlmutter DC Circuit decision (Mar 2025)