Photography, Generative AI, and the Recurring Anxiety of Technical Disruption
When photography emerged in the 1830s and 1840s, it did more than introduce a new image-making technique. It collapsed a long-standing arrangement in which skilled human labor—draftsmanship, likeness-making, representational fidelity—was scarce, slow, and valuable. A machine had entered the scene that could perform one of art’s most remunerated functions faster, cheaper, and with unsettling accuracy.
The reaction was immediate: anxiety, hostility, and a struggle over legitimacy.
Painters and critics questioned whether photography could be art at all. The objections were framed aesthetically and philosophically—mechanical process, absence of the human hand, lack of expressive intention—but they rested on a simpler material fact: photography threatened to devalue a set of trained skills that had previously underwritten both income and status. In that sense, resistance to photography shared a structural logic with Luddism.
The original Luddites were not merely enemies of progress: their grievances about skill, wages, and control over production were real, even if their tactics—smashing machines, sending death threats, and attacking local officials—were extreme and, at times, irrational. Their objection was not to technology per se, but to the social relations it reorganized. When 19th-century artists argued that photography cheapened realism or reduced it to a mechanical trick, they were making a Luddite claim: a machine was collapsing the value of a learned craft and threatening the livelihoods built upon it. The argument was aesthetic in form, but economic in substance.
This dynamic can occur even when the disruption is purely aesthetic. When abstract art emerged in the early 20th century, critics dismissed it for allegedly abandoning skill, technique, and expressive fidelity—claims strikingly similar to those leveled at painters facing photography. In both cases, the anxiety stemmed from the disruption of established hierarchies of labor, skill, and authority in the arts, whether triggered by machines or by formal experimentation. The form of the challenge changes, but the underlying social and economic tensions remain recognizable.
The parallels with generative AI are obvious—and not just in image-making. Generative systems now produce images, music, text, and video with minimal human input. In every domain, the same anxieties surface: job displacement, loss of craft, uncertainty about authorship, and fears that automated production hollows out meaning itself. Composers and musicians now hear arguments nearly identical to those once aimed at painters: that AI music lacks intention, that it dilutes skill, that it reduces years of training to algorithmic pastiche.
The medium changes: the structure of the anxiety does not.
History has already resolved one major part of this dispute. Mechanical or rule-based processes are not incompatible with art. Photography itself, along with found art, conceptual art, instruction-based work, and process-driven practices, settled that question decisively in the 20th century. The presence or absence of the hand is not a reliable criterion of artistic legitimacy. Whatever else generative AI may be, “it isn’t real art” is not a defensible objection on historical grounds.
Legal debates around authorship sharpen this analogy. In the 19th century, courts wrestled with whether photographs could be considered original works. Ultimately, it became accepted that a photographer could claim authorship by virtue of creative choices. The machine did not erase authorship: it reframed it.
History offers a warning against today’s reflexive panic. Photography did not destroy art. Film did not destroy theater. Recorded music did not destroy music-making. Craft persisted, mutated, and diversified. These facts matter, and they caution against delegitimizing generative AI aesthetically.
But historical precedent does not license complacency. The economic and institutional conditions are no longer comparable. The photography analogy holds for cultural reaction, but it breaks at the level of ownership, scale, and power. To assume benign outcomes simply because earlier disruptions were eventually absorbed is to mistake pattern for guarantee.
Generative AI will be integrated into creative practice. Some artists, writers, and musicians will use it as just another tool. Others will reject it on narrow ideological grounds. That much is inevitable. What is not inevitable is the shape of that integration, or who benefits from it. The comparison to photography should not reassure us that everything will work out. It should warn us that when machines move from being instruments—tools that extend the reach of the individual—to intermediaries—proprietary systems that sit between the creator and the canvas—the stakes change.
The question is not whether art will survive, but who gets to make it, and under what conditions.
Spunarounds 