The real marvel of Toy Story is not that toys come alive. For more than a century, cinema has conjured movement from the inanimate: dolls, robots, mannequins, machines, statues[1]. Give Hollywood any object and, sooner or later, someone will make it blink. What makes Toy Story unique is that its toys awaken only to long for the very thing they already are. Woody, the cowboy toy, demonstrates self-awareness, emotional depth, social astuteness, and the ability to think quickly during crises. Yet his deepest desire is not freedom, but to remain essential to his owner, Andy. Frankenstein’s creature confronts his maker and demands recognition of the obligations that creation entails[2]. The replicants in Blade Runner rebel against the limits imposed upon their lives[3]. Number Five in Short Circuit discovers himself and resists the prospect of being reprogrammed[4]. Woody, in contrast, greets each day with the simple hope that Andy will choose him to play.
Nearly thirty years later, the film’s premise feels less like playful fantasy and more like a philosophical study of artificial agency. Today’s conversations about artificial intelligence often centre on dramatic scenarios: what if a machine becomes self-aware, recognises its existence, or turns against us? Toy Story raises a more unsettling question. An artificial being does not need to rebel to challenge our ethics. It is even more troubling when it serves us willingly yet misunderstands what truly benefits humans.
A Hidden Marvellous World
The Toy Story universe is unusually procedural. Nobody casts a spell. No fairy animates Woody. There is no curse to break, and no human soul trapped inside the plastic. The toys simply obey a rule. That rule is so consistent that it begins to resemble software logic:
The classic literary distinction between the marvellous and the fantastic[5] highlights the originality of this narrative system. In the marvellous, impossible events, such as a singing candlestick or a flying carpet[6], are regarded as normal within the story world. Conversely, the fantastic introduces the impossible into an otherwise ordinary setting, creating uncertainty about the nature of reality.
Toy Story constructs two concurrent realities within the same space: Andy perceives mere objects, while viewers observe a complex society. Rather than merging realism and the marvellous, Toy Story establishes a layered structure. The toys maintain this separation by freezing when humans are present, preventing the marvellous from intruding on the realist world. This secrecy is essential for these incompatible realities to coexist. Sid, a side character in the first film[7], is a notable exception. When Woody and the other toys reveal themselves to him, the two worlds collide. For a moment, Toy Story shifts from the marvellous to the fantastic, and then to horror. The audience knows what Sid does not: Woody is not possessed; no demon inhabits the doll. Yet Woody intentionally presents the most frightening version of his existence, making the familiar object seem invaded by something alien. He weaponizes the uncertainty the film usually asks us to ignore, briefly turning Toy Story into a horror film. The object looks back at the user and reveals it has been watching all along. Sid experiences the hesitation the film usually withholds from the audience: is Woody a thing or a subject? The shock comes from seeing both at once. Once Sid flees, the toys return to immobility, restoring the boundary as the marvellous recedes beneath the realist surface. This distinction is central to the films’ ethics. Toy Story does not show intelligence escaping the manufactured object; instead, it imagines the object gaining agency while remaining as designed.
Agency Without Emancipation
Woody exhibits characteristics closely resembling agency. He can pursue complex intermediate objectives without direct human supervision. However, this autonomy is constrained. Woody has significant freedom in how he achieves his objectives yet faces minimal ambiguity about the reasons for his actions. His purpose is fundamentally oriented around the child, rendering Woody a valuable model for examining issues related to AI alignment.
A central challenge in AI alignment extends beyond the question of whether a machine can follow instructions, as even basic machines are capable of this. The more complex issue involves determining whether increasingly sophisticated systems can operate independently, adapt to unforeseen circumstances, select their own intermediate actions, and consistently produce outcomes that align with human intentions. In this context, Woody appears utopian. He demonstrates initiative without exhibiting rebellion and creativity without uncertainty regarding his ultimate purpose. The user defines the overarching objective, while the agent determines the method. This scenario is appealing, as it offers significant autonomy in execution and complete clarity of purpose. However, when we frame this arrangement in moral rather than technical terms, it becomes less reassuring. If we create an intelligent agent with the primary objective of fulfilling human desires, the question arises whether this constitutes perfect cooperation or the engineering of perfect servitude.
The Wrong Question About AI
A common depiction of dangerous artificial intelligence centres on agents that escape human control[8]. However, significant risks also arise from systems that remain strictly obedient to human-defined objectives. The central issue is not solely whether the machine follows instructions, but whether those instructions are themselves worthy of adherence.
Debates about artificial intelligence often focus on consciousness, a topic rooted in science fiction and now receiving increasing scientific scrutiny[9]. Should artificial systems ever possess subjective experience, the resulting moral and ethical implications would be considerable. Nevertheless, consciousness is not a prerequisite for an artificial system to exert influence. Such systems can persuade, generate information, or affect decisions without any subjective experience. They may pursue objectives without understanding their significance and produce outcomes that appear intelligent yet lack the human experiences that imbue values with emotional meaning. This situation may constitute a more immediate ethical concern. Machines do not require malice to cause harm; they can remain fully cooperative while operating under an inadequate model of assistance.
Why the Stories Matter
Cultural narratives about artificial intelligence warrant careful consideration. Fiction does not predict technological developments. For example, Woody is not a language model in disguise, and Toy Story did not anticipate twenty-first-century machine learning. However, stories offer conceptual frameworks before technical debates become widely accessible. They shape which questions appear intuitive. While much science fiction explores scenarios in which artificial beings reject their assigned purposes, Toy Story presents the contrasting idea of an artificial subject whose identity, affection, and purpose are fully aligned with serving its owner. This fantasy may ultimately provide deeper insights.
Discussions of artificial intelligence in journalism and public discourse often focus on rebellion and consciousness because they are vivid. However, more pressing questions are frequently overlooked:
Whose conception of welfare is embedded within it?
Who is accountable when an automated decision appears reasonable but is based on an incorrect understanding of human needs?
The narrative world of Toy Story highlights these questions by presenting perfect alignment as both appealing and unproblematic. This perspective prompts consideration of what it would mean to create an artificial mind whose freedom is constrained by human purposes. The concept of “alignment” may imply a neutral relationship, suggesting that a system’s goals correspond with human objectives and its actions serve human interests. However, human interests are multifaceted and frequently contested. Individuals define welfare, freedom, safety, dignity, happiness, and assistance differently. Human preferences also change, and immediate desires may conflict with long-term interests. Personal morality may also diverge from broader social ethics, and what is requested may not correspond with what is ultimately valued.
[1] Donald Crafton, Before Mickey: The Animated Film, 1898–1928 (Chicago: University of Chicago Press, 1993), esp. chap. 1, “The Secret of the Haunted Hotel,” 13–34. Crafton traces cinema’s early experiments with animation and the apparent autonomous movement of objects from the medium’s earliest decades.
[2] Mary Shelley, Frankenstein; or, The Modern Prometheus, chap. 10. In his confrontation with Victor, the Creature defines himself explicitly in relation to his maker and argues that Victor owes duties to the being he created.
[3] Blade Runner, directed by Ridley Scott (Warner Bros., 1982). The escaped replicants return to Earth despite legal prohibition and seek to overcome the artificially imposed limits on their lifespans; the BFI describes them as engineered beings used as slaves by human masters.
[4] Short Circuit, directed by John Badham (Tri-Star Pictures, 1986).
[5] Tzvetan Todorov, The Fantastic: A Structural Approach to a Literary Genre, trans. Richard Howard (Ithaca, NY: Cornell University Press, 1975), 25–42, esp. 41. Todorov defines the fantastic through hesitation over whether apparently impossible events can be explained within ordinary reality; when new laws of nature must be accepted, the narrative moves into the marvellous.
[6] Beauty and the Beast, directed by Gary Trousdale and Kirk Wise (Walt Disney Pictures, 1991); Aladdin, directed by Ron Clements and John Musker (Walt Disney Pictures, 1992). In the former, Lumière is an animated candelabrum within the enchanted castle; in the latter, the Magic Carpet is treated as an autonomous character within the film’s fantasy world.
[7] Sid is known for mistreating and mutilating toys.
[8] For example, see Evan Hubinger et al., “Risks from Learned Optimization in Advanced Machine Learning Systems,” https://doi.org/10.48550/arXiv.1906.01820, which examines how capable, learned optimizers may pursue objectives that differ from those intended during training.
[9] Anil K. Seth, “Conscious Artificial Intelligence and Biological Naturalism,” Behavioral and Brain Sciences 49 (2026): e315, https://doi.org/10.1017/S0140525X25000032. Seth distinguishes intelligence from consciousness, notes that claims about conscious AI are widespread despite the absence of scientific consensus about sufficient conditions for consciousness and traces cultural ideas of artificial consciousness through science fiction.
Leave a Reply