Around the time of the Cuban Missile Crisis, in an elaborate and costly operation, the cia flew objects over the island that were designed to look like alien spacecraft. In previous experiments with psychological warfare, ufos had been found to be one of the most effective consumers of mental bandwidth, driving targets to obsession and even insanity. The aim was to distract and unnerve the Cuban government and military: launched from a submarine, evading fighter interception and seeming to vanish in an instant, they served their purpose. This striking episode is one of many in Trevor Paglen’s stimulating new book about images made by, and very often for, machines. Pysops provide an unexpected analogy in his account of the contemporary image regime because, for Paglen, machine imagery is above all operational—intended to have an effect in the world, just as psychological warfare aims to manipulate the perceptions of the adversary.
Paglen is an unusual figure, a successful artist, writer and curator whose work has addressed secret military operations, state surveillance, and more recently machine-learning image algorithms and ai classification systems. Born in Maryland in 1974, he took an mfa at the Art Institute of Chicago and completed a PhD in Geography at Berkeley. Paglen is one of those artist-theorists whose practice informs their thinking and vice versa, and who have contributed some of the most acute analyses of the image world and its place within the wider system of capitalism and its states. We may also think of Allan Sekula, Coco Fusco, Martha Rosler and Hito Steyerl (it is no accident that work by the last two feature among the book’s illustrations). As with these other figures, Paglen’s writing oscillates between pure theorization and artist’s statement, which allows much greater freedom of association and expression in the former, and greater specificity and sophistication than is typical in the latter.
Paglen first gained prominence for aiming very long lenses, designed for astronomy, at secret military bases in the deserts of Nevada. Photographs taken from a mile away revealed scraps of information, such as the tail numbers on planes, but the greater the distance the more impressionistic the images became as aircraft, buildings and figures dissolved in heat shimmer and atmospheric haze. The photographs pointed towards something ordinary us citizens pay colossal sums for yet are forbidden to know about. Around the same time, he began writing a series of finely researched books, including Blank Spots on the Map: The Dark Geography of the Pentagon’s Secret World (2009) and with A.C. Thompson, Torture Taxi: On the Trail of the cia’ s Rendition Flights (2006), laying out what could be gleaned from the uniform badges of secret military units, erased areas in satellite ground imagery and the assiduous efforts of amateurs to track military satellites and the planes used by the us state to transport people it has kidnapped. He also contributed to the research and cinematography of Laura Poitras’s film about Edward Snowden, Citizenfour (2014).
As an artist, Paglen is known for making visually appealing, complex and cryptic photographic, graphic and sculptural objects, along with nfts. These include an ongoing photographic series tracking secret satellites, which takes the form of large-scale scenes of the evening or night sky, often seen above landscapes familiar from nineteenth-century photographs of the West made by Timothy O’Sullivan or Carleton Watkins. A winner of the MacArthur ‘genius’ fellowship in 2017, and this year of the lg Guggenheim Award, he is represented by one of the world’s most successful commercial galleries, Pace. This year, he co-curated the digital section of the major art fair, Art Basel. As is common with artists whose work is critical of the wider systems it is forced to inhabit, his work has a complicating double aspect—serving both to enlighten the ordinary gallery-goer and to decorate the homes of the super-rich.
Bringing together essays written over the past decade, How to See Like a Machine vividly illuminates its subject, and contains a strong argument about the unprecedented character of the contemporary image regime. Framed by a lucid introduction and conclusion, the first three chapters, originally published in 2016 and 2017, examine computer vision and its systems of image classification and identification; two further essays, from 2023 and 2024, then draw on the history of neurological experiments and psyops to find analogues for the cognitive tricks at play in these technologies, before a final essay turns to the ways that ai-generated imagery is altering the relationship between perception and reality. Illustrated throughout with a mix of archival and computer-generated images, these are captioned only at the end of the book, their enigmatic appearance imitating how humans today find themselves at sea interpreting today’s proliferation of images.
While the book is called How to See Like a Machine, it rapidly becomes evident that humans can do no such thing. Paglen tracks what he rightly sees as an epoch-making development, which is superficially marked by the recent inability of people to reliably distinguish machine images from those made by cameras or humans, and more profoundly the condition that has been reached in which ‘the overwhelming majority of images are now made by machines for other machines, with humans rarely in the loop.’ The computer file used to display a screen image may be thought of as a latent image, rather like an undeveloped photographic negative, unreadable to the human eye. Yet machines can read these files directly. Regardless of whether anyone attends to the two billion photographs uploaded each day to social media sites, machines scrutinize them for purposes of data mining, classification and manipulation.
There is little point, Paglen claims, in using old critical tools to analyse the meaning of such images. ‘Using traditional semiotic or iconological frameworks to think about image classifiers is like using art history to study military logistics’—we should ask instead how they act on the world. While the machines may operate on images in ways utterly alien to human attention, they are turned to all too familiar purposes: surveillance and advertising. Paglen highlights the striking example of social-media monopolies selling data from the images ‘shared’ on their sites to insurance companies, so that you can expect, if you post images of yourself enjoying extreme sports or on drunken evenings out, to pay more for your individual insurance policy. Little wonder that people are becoming increasingly cautious about what they post online and how they are photographed.