And why you can trust that the real ones are real.
Every real photograph and every Commons-sourced AI image here is public domain, CC0 or CC BY. Nothing under a share-alike licence, nothing scraped from a site that never agreed to it.
Share-alike is excluded for a mechanical reason: we crop and re-encode every picture, which counts as adapting it, and a share-alike licence would follow the adaptation onward. CC BY only asks for credit and for changes to be noted — you get both, under every picture, the moment you answer.
Every non-generated image was uploaded to Wikimedia Commons on or before 31 December 2020 — before any model could produce a convincing picture of anything.
It is a blunt rule and it throws away a lot of good photographs, but it is the only one we can actually prove. A game about spotting generated images has exactly one unforgivable bug: showing you a generated image and calling it real.
Real photographs — everything on Wikimedia Commons carrying one of these two licence templates, searched against a hand-written list of about 130 concrete subjects. A generated image is paired only with a real one filed under the same subject, and dropped outright if no subject matches:
Generated images:
Twenty additional AI images were generated with OpenAI’s GPT Image tool on 15 October 2026, specifically for Real or Not. The tool did not report its exact model version, so we label them GPT Image rather than claim an unverified version.
These images are provided for free use in this game by its publisher. This is not a claim that they carry a Commons public-domain or CC0 licence. Their real counterparts are CC0 photographs uploaded to Commons before 2021.
Generation records and prompts are retained privately. Both sides receive the same crop, colour normalization and WebP encoding. “Generated in 2026” records when we made the image; it does not claim a model release year.
The title, author, licence and a link to the source appear under both pictures the moment you answer — that is where credit belongs, and the only place we can put it without handing you the answer.
We do not publish one big list of every file used, for the same reason. Public domain and CC0 require no attribution at all, so nobody is owed anything by that choice; we show it anyway because it is decent, and because you should be able to check us.
Most real counterparts are photographs of places, nature and objects. To keep the scarce early-model material playable, a few legacy rounds use historical portraits of deceased people. Their source, photographer and licence appear after the answer.
The historical sources include Mark Twain (1835–1910), Nelson Mandela (1918–2013), and a photographic study from 1854. We crop the portraits to their subject; current people in the background are excluded.
Two legacy rounds compare AI-assisted graphics with human-made scientific graphics. Those real counterparts are labelled “a graphic”, rather than “a photograph”.
Every image is cropped to a square, stripped of metadata, given a random filename, pulled to a common colour saturation and re-encoded so that the two pictures in a pair end up a similar size on disk. These matching rules apply to both sides — and it is the change we are noting, for the images that ask us to note it.
It is not decoration. Before we did it, "pick the more colourful one" and "pick the smaller file" both found the generated image far more often than chance — two rules that beat an attentive human without looking at a single picture. If you find another one, tell us; that is the kind of bug that quietly ends a game like this.