Microsoft’s Tay chatbot provides an unusually effective opening case study because its rapid descent into racist and antisemitic output immediately undermines the idea that automated systems emerge independently from human choices, data, and culture. From there, the film makes a much larger argument: artificial intelligence should be understood not as an autonomous intelligence suddenly arriving in society, but as a collection of technologies embedded within histories of statistics, eugenics, military research, venture capital, labor exploitation, and concentrated corporate power. That framing gives the documentary a clear purpose, although it also creates an enormous burden of proof as it attempts to connect more than a century of intellectual and political history to the current AI boom.
The historical sections are among its most ambitious. Francis Galton, Karl Pearson, intelligence testing, the concept of general intelligence, forced sterilization, American eugenics, Nazi racial policy, Alan Turing, cybernetics, and early computer research are assembled into an argument about how efforts to quantify human ability influenced later ways of describing machine “intelligence.” The documentary is strongest when it identifies specific people, concepts, institutions, and intellectual relationships rather than treating technological development as an inevitable march toward smarter machines. Yet lineage is sometimes treated too readily as causation: demonstrating that statistical methods were developed or used within eugenic projects does not by itself establish that modern machine-learning applications employing related mathematics inherit the political purposes for which earlier researchers used those tools.
That tension becomes even more pronounced as the story reaches Silicon Valley. William Shockley’s semiconductor legacy and his openly eugenic arguments make him a striking figure through whom to examine the uncomfortable coexistence of technological achievement and racist ideology. The film then expands this into a critique of the Valley’s mythology of exceptional male founders, technological inevitability, libertarianism, and the recurring promise that innovation will solve problems traditionally addressed through politics or collective institutions. Archival footage works particularly well here, allowing past predictions about computers surpassing humanity to echo modern claims about AGI and the singularity. The weakness is that provocative phrases such as “techno-fascism,” along with broader claims about the ideological continuity of Silicon Valley, sometimes receive more assertion than sustained examination of competing interpretations.
The chapters on AGI, superintelligence, effective altruism, and the singularity provide one of the film’s sharpest critiques of AI rhetoric. Rather than accepting AGI as a clearly defined technological milestone, the documentary emphasizes disagreements over what intelligence and AGI even mean while showing executives and theorists making extraordinary predictions about human-level machines, intelligence explosions, and transformative futures. It persuasively demonstrates that speculative ideas have become influential within technology investment and corporate strategy, but its attempt to trace the “G” in AGI directly through psychometrics and eugenics is more interpretive than conclusively demonstrated here. Similarly, characterizing parts of the AI-safety movement as cult-like is rhetorically memorable but receives less rigorous support than the film’s narrower criticism that speculative existential risks can command enormous attention and funding.
The shift from intellectual history to present-day labor makes the critique substantially more concrete. A Kenyan worker describes participating in AI training work for less than a dollar per hour and encountering graphic material that allegedly produced serious psychological consequences, while the broader discussion argues that supposedly automated systems remain dependent on large amounts of poorly visible human labor. Whatever one makes of the film’s larger description of this arrangement as “digital colonialism,” the worker testimony effectively challenges the popular image of AI as an immaterial machine performing everything by itself. The subsequent examination of data centers similarly redirects attention from futuristic language toward physical infrastructure, energy, water, land, chips, and communities asked to accommodate enormous computing facilities.
By its final chapters, the scope expands again into government procurement, the Trump administration’s AI policies, military contracting, synthetic media, information integrity, corporate political influence, and the concentration of technological power. The film presents these developments as evidence that AI is becoming intertwined with state and military power rather than remaining primarily a consumer technology. Its excerpts concerning federal procurement and ideological neutrality reflect policies publicly advanced by the Trump administration, including requirements aimed at obtaining models described by the administration as objective and free from ideological bias. But the documentary frequently moves from documented policy or corporate behavior into much broader conclusions about fascism, colonialism, militarization, and elite intentions without consistently separating what has been established from what its interviewees infer those developments ultimately mean.
Pros
- Uses Tay as an effective opening example of how supposedly autonomous AI output can reflect human data, design choices, and social behavior.
- Builds an unusually broad historical account connecting debates over intelligence, statistics, computing, Silicon Valley culture, and contemporary AI.
- Makes effective use of archival predictions about intelligent machines to place current AGI and singularity rhetoric in historical perspective.
- The Kenyan data-worker testimony grounds the discussion of AI labor in specific human experiences rather than leaving exploitation as an abstract claim.
- Examining data centers, water, energy, capital, and human labor usefully challenges the popular conception of AI as an essentially immaterial technology.
- Regularly questions vague terms such as AI, AGI, intelligence, and superintelligence instead of accepting industry terminology at face value.
Cons
- The enormous historical argument sometimes treats intellectual association or mathematical lineage as stronger evidence of ideological continuity than the material presented can establish.
- Connections between eugenics, psychometrics, general intelligence, and modern AGI are provocative but occasionally compressed enough that important distinctions between concepts and applications receive limited attention.
- Labels including “techno-fascism,” “cult,” and “digital colonialism” carry substantial analytical weight without always receiving equally substantial examination of alternative interpretations.
- The overwhelmingly critical selection of experts and examples gives comparatively little room to researchers or practitioners who could challenge the film’s historical and technical arguments rather than merely represent industry promotional rhetoric.
- Later sections combine corporate economics, environmental costs, military contracting, Trump administration policy, synthetic media, labor exploitation, and democratic concerns so rapidly that some potentially strong arguments receive less development than they deserve.
An impressive combination of technological history, labor reporting, archival material, and ideological critique turns familiar AI debates away from hypothetical sentient machines and toward the humans, institutions, resources, and political choices already shaping the technology. Its strongest evidence raises serious questions worth considering independently of its more sweeping historical connections, while its tendency to present disputed interpretations with limited counterargument keeps an ambitious investigation from being fully persuasive.
