The Slow Death of Images: Where Do Our Digital Photographs Disappear To?
26 • 03 • 27Liliom Szabó
Floppy disks, cassettes, CDs, and DVDs—data carriers that once embodied the promise of a modern future—are now primarily found in antique shops. Today, we rarely store our images on physical media; instead, we subscribe to cloud-based services, believing that we can access our memories anytime, anywhere. What do we lose by preserving our images as pixels? Where do photographs taken years ago disappear to—do they fade into oblivion? What exactly is the cloud, and what impact does it have on the environment?
For many, cloud services seem like a mystical space where photographs ultimately end up. On seemingly well-organized platforms, users can retrieve images by date and location, arrange them into albums, and share them freely. Thanks to facial recognition technologies, we can even search through thousands of images for a specific person—or even an animal. Meanwhile, we live with the assumption that all we need is an internet connection to access any of our pictures from any device, from that undefined place we call “the cloud.” However, our images are not floating in the ether; they reside within the physical infrastructure of vast, energy-intensive server farms. In Europe, eight such facilities operate, including in the Netherlands, Belgium, and Denmark.
As Blair Spowart, co-owner of Seedling, a platform focused on reducing carbon footprints, notes, the environmental impact of this infrastructure is significant. The global digital sector, including data centers, networks, and user devices, is estimated to account for roughly 3–4% of worldwide greenhouse-gas emissions. Data centers alone consume around 2% of global electricity—approximately as much as some entire countries—and their energy demand doubled between 2015 and 2022. As cloud-based services continue to expand, the emissions associated with digital storage are becoming an increasingly significant part of the climate crisis.[1]
Depending on where we store our photographs, we may not even be sure that the company will still exist in 20 years. Another question concerns what will happen to our digitally stored images once we are no longer here. Who will sift through accounts containing thousands of images to retrieve the visual evidence of our lives? We tend to treat giants like Google as if they were as permanent as the sun or the moon. Yet even here, uncertainties arise. Storing data—photographs, in this case—in public clouds means relinquishing a certain degree of control over access and management. We trust that the cloud provider will always be able to make the data accessible and maintain its systems and data security. Although public cloud providers strive for constant availability, outages do occasionally occur, rendering stored data temporarily inaccessible, as we can read on Google’s own website.
The dilemmas of environmental impact or unpredictable access primarily address the practical aspects of digitally stored images. Yet the topic also invites aesthetic and cultural-anthropological reflection. The first known photograph in human history was taken from the French apartment of Joseph Nicéphore Niépce and is still carefully preserved today in a museum in Texas. In the analog era, the original image carried immense significance. But as Fred Ritchin notes in After Photography (2009), the digital image becomes “an abstraction in which the original and the copy are no longer distinguishable; the original loses its meaning.”
We can produce infinite variations in limitless quantities, yet most of them will never reach physical reality. Ritchin argues that digital images are no longer merely images but networks of information, hyperlinks, and contexts. This shift points to a stronger, multidimensional visual language capable of connecting images, events, and ideas. While this opens new levels of interpretation, it simultaneously further diminishes our sense of possession over photographs. As Marvin Heiferman writes in Photography Changes Everything, “As photography transforms, so too does the tacit contract between images, reality, and viewers.”
0103
“Photograph of Abraham Lincoln taking a selfie.” This is a synthetic image, not a photograph, generated by DALL·E in response to the text prompt (above) by Fred Ritchin, February 2023.
Where carefully selected negatives were once arranged into albums or archives, today the amateur image-maker—most often photographing with a phone—is left with cloud-generated thematic compilations, “decorated” with background music, which are often not even searchable afterward. Instead, algorithms generate a “personal” experience for us on a given day. A recent study reported by Helena Horton in The Guardian reveals that most of the images we store in the cloud are never opened again. These include screenshots or memes that cannot truly be considered photographs. Such data occupies space in server farms, quietly, without our remembering them.
Yet valuable images can also be lost in these processes. Regardless of genre, we store them all in the same place, and as the volume of data grows, retrieving truly meaningful photographs becomes increasingly complicated and exhausting. Thus, although more photographs are being taken now than at any other time in human history, we may be the generation that leaves behind the least visual evidence in physical form.
Perhaps we can occasionally review our images in the cloud and delete those we no longer need—or ever needed. We should also print—or otherwise materialize—the images that deserve it, ensuring that their survival a hundred years from now does not depend solely on the continued existence of a server farm, so that future generations may still look back and see how people once lived.
[1] Spowart, Blair. “Cloud Emissions: The Ultimate Guide to the Carbon Footprint of the Cloud in 2025.” Seedling, 9 Jan. 2026, https://seedling.earth. Accessed 28 Jan. 2026.
