This page explains how Artificial Consequences chooses titles, scores them and corrects them.
What we review
We cover films, TV series, anime and games in which an artificial intelligence or automated system matters to the story. Reader suggestions are one of the ways titles reach the index.
How scores work
Each title is scored from 0 to 10 in four equally weighted categories: technical feasibility, human behaviour, institutional response and causal consistency. The Realism Index is the average of the four, rounded to one decimal place. The full rules are on the method page. Scores are editorial judgments, not measurements, and each case file explains the reasoning behind them.
Sources
Every case file names the work it covers and links to a reference listing for it, such as its IMDb page. The interpretation and scores are our own.
Spoilers
Case files discuss plots in full, including endings. Each one carries a spoiler note above the analysis.
Illustrations
Case files use original editorial illustrations made for this site.
Revisions and corrections
A score can change when new evidence emerges, when a story continues, or when we find a mistake in our reasoning. When that happens, the case file is updated with a dated note explaining what changed and why. Factual errors are corrected as soon as we confirm them.
If you think something is wrong, tell us through the suggestion form and name the case file in your message.
Independence
We do not accept payment, free access or other incentives in exchange for coverage or scores, and the site is not affiliated with any studio, network, publisher or game developer.