From plotter drawings in the 1960s to images conjured from a text prompt today, digital art is less a single style than a way of working. Here is what the term really means, and how to begin.

When we call something digital art, we are describing not a subject or a style but a set of tools and a way of making. Digital art is any art created or presented using digital technology, whether it lives on a screen, spills out of a projector, drives an interactive installation, or is registered on a blockchain. A picture built pixel by pixel in software, a short film of shapes that generate themselves, a website conceived to be experienced as art, an image summoned from a line of typed text: all of these sit under the same broad umbrella. What unites them is not how they look but the machine at their heart.

What makes an artwork digital

The defining trait is simple. In digital art the computer is both the medium and the tool, the canvas and the brush at once. An oil painter works pigment into linen; a digital artist works numbers into pixels, and those numbers can be copied, altered, and moved without any loss. That leads to a second quality that unsettles older ideas about art, which is reproducibility. A digital file has no single original in the way a marble carving does. It can be shown on a thousand screens at once, each one as faithful as the last, which forces us to ask where the artwork actually resides.

Two further qualities complete the picture. Many digital works are interactive, changing in response to a viewer who clicks, moves, or simply walks past a sensor, so the audience becomes a kind of collaborator rather than a spectator. And underneath a great deal of digital art sits code, written instructions that behave less like a finished object and more like a recipe. When the code is the true material, the artist composes a system and then lets it run, and the image on the screen is only one of countless outcomes it could produce.

A short history from plotters to prompts

The story begins earlier than most people expect. In the 1960s a handful of scientists and artists began coaxing images from room-sized computers, feeding instructions to mechanical plotters that drew crisp geometric patterns in ink. Much of this pioneering work was gathered in the landmark exhibition Cybernetic Serendipity, staged at the Institute of Contemporary Arts in London in 1968, which introduced a wide public to the idea that a machine could be a partner in creation. For two decades the field stayed largely in the laboratory. Then, in 1990, the release of Photoshop and the arrival of affordable personal computers put powerful image-making software on ordinary desks, and digital art quietly became something anyone might attempt.

From there the pace quickened. In the 1990s a generation of net artists treated the young World Wide Web itself as their canvas, making pieces that lived only in the browser and could not be hung on a wall. Alongside them, artists working in the older tradition of generative and algorithmic art refined the practice of writing programs that produce endlessly varied results. In 2021 the wider world caught up in dramatic fashion when the digital artist Beeple sold a collage as a non-fungible token, or NFT, at Christie’s for some 69 million dollars, and for a giddy season the market believed it had found a way to make files scarce and collectible. The speculative bubble soon deflated, yet the underlying idea of verifiable digital ownership did not vanish. Today the conversation has shifted again to artificial intelligence, whose image generators can produce finished pictures from a written description in seconds. It is a genuinely powerful tool and a genuinely contested one, raising fair questions about the artists whose work trained it, and it is best approached with curiosity rather than either hype or alarm.

The many forms it takes

Because the medium is so flexible, digital art has branched into many distinct practices. Digital painting and illustration are the most familiar, made with a stylus and tablet in programs that mimic and then far exceed the behaviour of real paint. Three-dimensional modelling and motion graphics build sculpted worlds and moving images for film, games, and animation. Generative and algorithmic work lets code do the drawing, while glitch art does the opposite, treating errors and corrupted files as raw beauty. Net art continues to use the browser as its gallery, and interactive and installation pieces fill physical rooms with responsive light and sound. Newest of all is AI-assisted work, in which the artist directs a model through prompts and careful editing.

These categories are porous, and they spill readily into the world of design. The same tablet and software an illustrator uses for a gallery print also produce logos, posters, and packaging, which is where digital art shades into applied craft. If that overlap interests you, our guide to what is 2D applied art traces the long tradition of art made to be useful, a lineage that today’s digital designers extend with every screen they touch.

Why it matters

There is a strong case that digital art is the most widely made and most widely seen art of our time. Every day, more images are drawn, rendered, filtered, and shared on devices than were painted by hand in whole centuries, and the phone in your pocket is both museum and studio. Beyond scale, it presses on the oldest questions in art with fresh urgency. If a work can be copied perfectly and endlessly, what do we mean by an original? If a program or a trained model generates the final image, where does authorship lie, with the coder, the tool, or the countless hands whose work fed the system? These are not idle puzzles. They touch on value, ownership, and credit, and how we answer them will shape how artists are paid and remembered.

How to start

The barrier to entry has never been lower. Capable software now spans the full range from free to professional, so a beginner can open Krita or GIMP without spending a penny, then graduate to tools such as Photoshop or Procreate when the work demands it. A pen tablet, which lets you draw directly with a stylus rather than a mouse, is the single most transformative piece of hardware and can be found at every price. If the coding side calls to you, friendly creative languages such as Processing and its browser-based cousin p5.js were built expressly to teach artists to make images with instructions, and both are free and generously documented. Most valuable of all may be company. Online communities gather to share work and swap techniques, and nothing sharpens a new practice faster than showing it to others. Pick one tool, make something small and finish it, and let the medium teach you the rest.