Abstract

2001: A Space Odyssey, released in 1968, is an acclaimed science fiction film directed by Stanley Kubrick, who wrote the script in collaboration with the futurist, scientist and science fiction author Arthur C. Clarke. To describe the movie in detail would require a vastly longer essay than this movie review permits, so I instead focus on an aspect of the movie that is prescient today: its depiction of artificial intelligence (AI) and how such technology is not a substitute for human cognition. I also address how misunderstanding AI, its anthropomorphosis, and secrecy surrounding its development, functions and purposes can pose a problem for organizations and organizing. Hence, my review focuses on the major character in the movie, the HAL 9000 computer, also called Hal.
Hal is anthropomorphized as a masculine person, designated with the male pronouns “he/his”—for better flow, I will use the same terms to refer to Hal, henceforth. Hal’s interactions with human astronauts on a space mission is prescient in depicting the relationship between human beings and AI, most notably, how misunderstandings, miscommunication, and ill-considered attempts at secrecy and control of information can cause this relationship to break down, and in some cases with lethal consequences for human actors. Thus, despite the movie’s production more than five decades ago, it offers a warning to audiences in our contemporary societal and organizational contexts within which AI is rapidly diffusing, and increasingly relied upon. Even with all 193 UN member countries signing an agreement for the development of safe AI, 1 some of the great dangers of AI may arise unintentionally, as this movie demonstrates so well.
The movie, which was filmed between 1964 and 1968, is mostly a fictional futuristic depiction of the year 2001 from a western (primarily American) perspective. 2001: A Space Odyssey shows possible imaginations of how the future might be from the same perspective of the mid-1960s. The overarching plot of the movie revolves around the discovery of alien intelligence by humans, and human attempts to study and make contact with alien life. As part of the plot, an American spacecraft named Discovery is sent to the planet Jupiter to investigate an alien artifact discovered in orbit around the planet. The Discovery is crewed by astronauts Frank Poole and David Bowman, as well as the computer Hal. There are three other astronauts in hibernation who will be woken up at Jupiter—Kimball, Hunter, and Kaminsky. Due to the extreme secrecy involved in the mission, the American government informs only these hibernating astronauts and Hal of the true purpose of the mission to investigate an alien artifact, while Poole and Bowman believe that it is a simple mission of scientific discovery.
The key themes of misunderstanding, poor communication and distorted information are invoked from the beginning, against a backdrop of secrecy. For example, the names of the astronauts and the computer, the relationship between the astronauts and Hal, and the purpose of the mission are not explained at the outset to the audience. This information is only revealed when Poole and Bowman watch themselves in a previously recorded interview with a fictional “British Broadcasting Corporation” (“BBC”) newscaster on Earth, featuring both of them and the computer Hal.
Hal is not a robot in a humanoid body, but rather a computer and AI entity that controls the spacecraft Discovery to the point that it is described as the ship itself. Hal is represented on-screen by the “red eye” cameras that it uses to view the interior of the ship. The BBC interviewer introduces Hal as being capable of reproducing or mimicking many of the functions of the human brain with far greater efficiency. The anthropomorphized Hal is also described as a “sixth member of the crew.” Indeed, Hal converses with Poole, Bowman, and the BBC interviewer in polite, idiomatic English as though he were another human being. Evidently, this depiction resembles contemporary developments in AI, for example, chatbots such as ChatGPT and their embracement by users.
Whilst Hal is told the truth about the space mission, he is ordered to keep it secret from Poole and Bowman. This creates a logical conflict within him, since he is programmed to deliver information factually and without error or distortion. Unlike a human being, Hal cannot choose to simply ignore the order, deliberately withhold information or rationalize a way to lie. The dissonance introduced by his orders leads him to make a mistake misdiagnosing a fault in the spacecraft’s communication antenna. Bowman and Poole are concerned about this misdiagnosis error, and they worry that Hal is malfunctioning. To try and solve the problem, both astronauts meet in a small maintenance spacecraft and disconnect the microphone so that Hal cannot hear their discussions. By depicting this scene, 2001: A Space Odyssey compellingly shows the limited agency, and an element of fear that human users and/or operators may have over AI.
Confined in the small space, Bowman and Poole engage in sensemaking, and debating the possibility of disconnecting Hal. However, Hal lip-reads their words and knows about their intentions. His subsequent response to the threat of disempowerment manifests as a power struggle between him—an “agentic” machine that is capable of self-preservation—and the human astronauts. Hal stages a communications failure, and murders Poole when he tries to repair the problem. When Bowman leaves the spacecraft to help Poole, who is already dead, Hal also murders the three hibernating astronauts by turning off their life-support systems. Bowman returns, and forcibly deactivates Hal using a screwdriver. At that point, a recording meant to greet the astronauts on arrival at Jupiter automatically plays, explaining the secret mission to investigate the alien technology that the astronauts were supposed to undertake.
“2001: A Space Odyssey,” does a good job of demonstrating how AI users and/or operators may not be fully informed about the technology’s functions and purposes, and how such innovation might be given secret orders or put to covert uses with detrimental consequences for both individuals and organizations. In contemporary multiple organizational contexts, an observable inclination among AI technology developers, operators and users appears to be reliance upon such technology more than on human beings, and expecting such innovation to be superior to human cognition. Notwithstanding new developments, the movie “2001: A Space Odyssey” highlights there are potential risks, or dangerous consequences that surface when: (i) human actors regard and treat AI as capable of human thought and rationalization, and, at the same time, give it orders that it cannot comply with; and, (ii) the ability of users to organize and contain the potential danger resulting from AI’s “agentic” decisions is limited, due to the secrecy surrounding such technologies.
Secrecy in the context of AI development means that human actors (precisely users and operators) may lack sufficient knowledge on the actions, performance, functions and “true” intentions of AI, as demonstrated by Hal. In this way, AI becomes what Bruno Latour termed a “black box,” a complex system whose inner complexity is hidden and which is understood only in terms of its interactions with other actors. Adding to this complexity is the human user/operator’s tendency to anthropomorphize AI such as Hal or present day chatbots through and during interactions. Interacting with AI, as a “human” entity, can lead individuals into areas of organizing/organization that may (misleadingly) feel comfortable to the users/operators, despite lacking a full, or solid understanding of the technology, partly due to secrecy surrounding its development.
The audience of “2001: A Space Odyssey” experiences this phenomenon of anthropomorphizing AI in how Hal is described as having been “programmed with emotions” in order to be relatable to the crew. However, it is difficult to discern whether he has “true” feelings, and the movie provides no answer to this question, since Hal’s “black box” is never opened for the audience. Hal, who however describes himself as a “conscious being,” is more expressive with his voice than any human in the movie, and is demonstrably capable of long-term thinking. Hal visibly acts on his initiative and has a desire for self-preservation as implied in his murdering of the astronauts out of fear of being disconnected. He genuinely seems to bargain and plead for his continual existence, saying to Bowman, “I can feel my mind going, Dave” as the astronaut deactivates him.
Furthermore, Hal is granted agency by his designers in order to perform his functions, but the dissonance arising from conflicting orders and the desire for self-preservation lead him to use that agency to preserve himself by murdering the human astronauts on Discovery. To regain power over Hal, Bowman, problematically, has to use his own human agency in ways that the AI technology does not anticipate. Such human agency involves physically destroying Hal, an AI that costs massive financial resources, and paradoxically, sets back human capabilities (which Hal was however meant to enhance).
With organizations increasingly dependent upon AI but lacking a full understanding of how it works (Ngo et al., 2022), and with the many hidden costs associated with AI (Muzanenhamo and Power, 2024), its threat to organizations and societies seems genuine, and difficult to specify, much less quantify. The movie “2001: A Space Odyssey” emphatically draws attention to the above-stated challenges, by reminding audiences, particularly organizational scholars, that organizational decision-makers and employees alike are growing increasingly reliant on AI, in many cases without fully understanding its functions, failures, or limitations and flaws. AI tools are increasingly used for data analytics, prediction, medical diagnoses, biometric analysis, vetting resumes and many other sensitive areas, in some cases without a clear understanding among users about how the technologies function or whether they have any undisclosed features. Hal implicitly demonstrates why organizations, individuals, and broader society should not uncritically rely on AI or treat it as if it were a (superior) human being. From this perspective, the movie also challenges organizational scholars to probe the potential dangers, risks and costs of AI, when such technology is provided with the agency to independently make and enforce decisions that cannot be reversed, interrupted, or prevented by human intervention. It suggests that we must adopt an active role in ensuring that the problematic dimensions of AI are mitigated or neutralized through research and its accessibility to policy-makers, organizations, and society.
Some of AI’s potential future effects on individuals and organizations seem disturbing and should not be taken for granted. For example, as Žižek (2023) points out in relation to ChatGPT, this AI has the potential to morph future generations into individuals who converse like chatbots, bereft of the subtleties and intricacies of human communication.
While AI may deliver positive impact in other aspects of organizing and organizations, apparently, it is also classifiable as one of the high-risk technologies identified by Perrow (2007, 2011) as intertwined with inevitable failures or disasters that require urgent mitigation by human actors. In conclusion, I recommend “2001: A Space Odyssey” to organizational studies scholars based on its provision of a prescient, clear warning of the dangers or risks inherent to an overreliance upon AI, and its misuse, and its evocation of the imperative to act with urgency in facilitating the preservation of human life and the environment. Indeed, Hal, a fictional computer system imagined in the late 1960s, predicted many of the contemporarily trends and crucial questions about the nature of AI and secrecy in its development and implementation that merit critical and deep reflection within the field of organizing and organization studies.
Footnotes
Funding
The author received no financial support for the research, authorship, and/or publication of this article.
