“The old world is dying, and the new world struggles to be born; now is the time of monsters.” Italian philosopher Antonio Gramsci’s line opens “Ghost in the Machine,” a Sundance documentary directed by independent filmmaker Valerie Veatch that uses its 110-minute runtime to ask viewers to reconsider what exactly feels “new” about artificial intelligence. The film firmly positions AI as the continuation of a conveniently forgotten intellectual and ideological history rather than as an abrupt technological innovation.
Through a collage of nearly 40 interviews intertwined with archival material spanning decades, Veatch traces this lineage backward through the institutional forces that made AI’s emergence possible. She even mentions Dartmouth itself and the 1956 Dartmouth Summer Research Project on Artificial Intelligence, where computer scientist John McCarthy first coined the term “artificial intelligence.” She pulls footage from a public debate in which McCarthy reflected on this moment with disarming candor, admitting that he “invented it because we had to do something when we were trying to get money for a summer study.” This clip reinforces the documentary’s claim that AI, even at the level of its naming, has always been entangled with funding pressures and the cultivation of excitement around technological possibility.
“Ghost in the Machine” contextualizes artificial intelligence within earlier philosophical debates about mind, intelligence and mechanization. The film notably invokes British philosopher Gilbert Ryle’s rejection of Cartesian dualism — René Descartes’ theory that an immaterial mind exists separately from and controls the physical body — the original “ghost in the machine.” Ryle coined the documentary’s titular phrase to mock the belief that an invisible mind operated inside an otherwise rational system. This concept becomes the groundwork for the documentary’s central argument that artificial intelligence is fundamentally human in origin. Veatch reworks this “ghost” into something far more unsettling and relevant to contemporary AI, suggesting that systems often presented as mechanical and neutral remain animated by unseen human forces, from embedded values to inherited biases.
From here, “Ghost in the Machine” turns its focus toward the ideological foundations of Silicon Valley. Veatch complicates the familiar modernization narrative of technological advancement by revisiting figures such as William Shockley, who was widely celebrated as a foundational innovator and co-inventor of the transistor but also one of the most outspoken proponents of eugenic thinking within mid-century American science. The juxtaposition reveals how narratives of efficiency and optimization have long coexisted with hierarchical assumptions about intelligence and human value.
This same logic resurfaces through the film’s examination of the 2023 paper released by a team of researchers affiliated with Microsoft and OpenAI, “Sparks of Artificial General Intelligence: Early Experiments with GPT-4.” Before the original authors revised it due to public backlash, the paper originally cited a definition of intelligence from psychologist Linda Gottfredson, whose work has been funded by the Pioneer Fund, a foundation long criticized for supporting research tied to scientific racism and hereditarian intelligence theories. By pointing this out, Veatch forces viewers to reckon with the uneasy possibility that definitions once used to rank human intelligence are now shaping how machine intelligence is understood, echoing histories that feel both dehumanizing and uncomfortably present.
The documentary’s most impactful moments emerge when it shifts focus to the human labor that sustains AI. In Nairobi, described within the film as the “Silicon Savannah,” data workers recount performing content moderation and annotation tasks essential to training machine learning systems while earning wages that scarcely reflect the value of their work. Particularly compelling is the film’s investigative reporting on the hiring practices of Samasource, Meta’s outsourcing provider, including evidence that applicants are funneled through application interfaces that restrict selectable locations to specific lower-income communities, a targeting practice that takes advantage of the economically vulnerable and enables low wages to persist. This revelation underscores a sobering throughline within “Ghost in the Machine”: Artificial intelligence may be marketed as a universal technological advancement, yet its development continues to rely on deeply uneven distributions of power and opportunity.
Although “Ghost in the Machine” occasionally risks stating what many viewers may already suspect about the rapid expansion of AI and unchecked dominance of tech giants, that familiarity ultimately becomes part of its force. Through its layering of historical context, its gradual turn toward critiquing modern systems of institutional power and a wide range of expert voices that steadily draw connections for the viewer, the film produces a cumulative, connect-the-dots structure. It ultimately poses a question that feels difficult to dismiss: If the systems of influence, economic incentives and labor inequalities shaping artificial intelligence are increasingly visible, then why does meaningful intervention — whether in the form of regulatory scrutiny, labor protections or shifts in public accountability — remain so elusive?
Veatch’s decision to prioritize accessibility over technical deep dives into model architectures allows the documentary to resonate across audiences — beyond specialist communities. Supported by a tech-adjacent synth score and an array of striking visuals — a mix of AI-generated imagery, live-action footage and stock footage that occasionally veers toward arbitrary — “Ghost in the Machine” emerges less as a revelation than a welcome reframing. By recovering the forms of humanness embedded within artificial intelligence, including dominant power structures, ideological commitments and invisible labor, the film insists that AI is neither autonomous nor inevitable, but rather constructed and contingent. In doing so, it leaves viewers not with answers, but with a sharpened awareness that the AI systems it interrogates are not distant technologies. They exist within our present world. More importantly, the film suggests that because AI remains fundamentally human, its future is not predetermined, leaving room for refusal and collective efforts to shape more equitable technological futures.
