A technology project that used facial compositing software to merge features from multiple celebrities deemed conventionally beautiful produced a single "ideal" face that generated significant discussion — and not all of it flattering to the premise.

The project drew on a tradition in psychological research on attractiveness that dates to the work of Francis Galton in the 19th century. Galton discovered that composite photographs — images created by overlaying multiple individual faces — were typically rated as more attractive than most of the individual faces that composed them. This observation, which has been replicated many times since, is thought to reflect the mathematical averaging of features: composites are closer to population averages, and average facial proportions are consistently rated as attractive across cultures.

The celebrity composite project took this principle and applied it to a curated selection of women widely considered beautiful: a mix of Western and non-Western faces, actresses, and models. The resulting image was symmetric, smooth-featured, and notably — if predictably — indistinct.

Critics observed that the exercise inevitably reflected the biases of whoever selected the source images. The definition of "beautiful" that the algorithm was asked to average was already the product of a particular cultural moment's preferences. The output tells us less about some universal standard of beauty than about which faces were most prominent in a specific media landscape at a specific time.

The deeper question raised by such projects is whether beauty is better understood as a fixed property of faces or as a shifting social construction that technologies can describe but not define. The composite face is interesting. It is not, as some claimed, objectively the most beautiful woman in the world.