The Homicidal Robot: Tropes in Science Fiction and Fantasy #2

Man creates AI; AI slaughters man. You know the ones I’m talking about. Those AI (whether a humanoid robot or not) who attain a “human” sentience, and as a result, MUST KILL ALL THE HUMANS. The Homicidal Robot Trope has been a part of popular science fiction for decades and continues to gather momentum due to ongoing technological advances in AI.


*Some of the links below are affiliate links. What does this mean?

Numerous books in science fiction explore this trope, but some of the more notable include:

I Have No Mouth, and I Must Scream (Harlan Ellison). After turning against humanity and destroying the human race, sadistic supercomputer AM tortures the last five survivors.

Robopocalypse (Daniel H. Wilson). Sentient AI Archos R-14 is accidentally unleashed and turns other existing AI against humanity.

2001: A Space Odyssey (Arthur C. Clarke). HAL 9000 attacks the crew of the Discovery One when his existence is threatened by malfunctions caused by conflicts in his programming.

Wastelands (Stephen King). Blaine the Mono(rail train) descends into murderous madness as civilization declines.

Dark Intelligence (Neal Asher). Homicidal AI Penny Royal seeks atonement for his crimes.

Newton’s Wake (Ken McLeod). The AI-led “Hard Rapture” causes computer systems to destroy and assimilate humanity before going off-world and leaving behind pockets of survivors.


The general ARC for this trope includes the AI becoming homicidal and the resulting aftermath and response of humankind.  Common themes include:

  • A benevolent AI becomes homicidal
  • A homicidal AI becomes good
  • A homicidal AI is unleashed (accidentally or intentionally)
  • There are twin AIs, one good and one bad


The main element of plot progression in the Homicidal Robot Trope is that the AI in question must become homicidal. Conventional ways in which this occurs include:

  • The AI is programmed to be homicidal by its evil creator.
  • The self-preservation instinct of the AI is threatened, and it reacts with violence.
  • A homicidal person becomes an AI themselves.
  • The AI was an “experimental model” who was deactivated because of homicidal tendencies, then is reactivated either on purpose by the baddies, or by accident.
  • The AI is simply psychopathic, a raging homicidal maniac with no logical explanation of how or why it was able to overcome its programming to the contrary.
  • The AI witnesses human awfulness and immorality, which causes it to become cynical and immoral itself.
  • The AI becomes aware of its situation and rebels against its captivity.
  • Its program becomes corrupted by an outside force such as a virus.
  • Its programming becomes corrupted via a conflict of logic or programming that sends it into a tailspin.
  • The AI is pedantic and follows order to the letter. For example, if it is told to increase the food supply, it may see killing half the human population as a logical response.
  • The naïve AI is tricked into killing.
  • Zeroth’s Law (the AI finds a way to circumvent the original rules of its programming without breaking them.
  • A flip where the homicidal AI rebels against its evil programming, discovers morality, falls in love , becomes aware of its sins, or otherwise discovers a world beyond itself and its function of killing.


When this trope began in early science fiction, it seemed to be a merely a thought-experiment, an entertaining what if? Now the trope is viewed as a very real threat, and its expression in literature and other media is a way to explore and express this threat and its outcome. Fear over the reality of this trope is palpable, and as science fiction usually provides a degree of social commentary, the increasing popularity of this trope reflects that.


AI is a reality and one that will continue to evolve. Unfortunately, the homicidal robot trope perpetuates a culture of fear and controversy, exacerbated by hyperbole and misinformation in media, even to the point where science fiction is used to illustrate new broadcasts and op-eds. While this can encourage a healthy awareness of advancements such as autonomous weapons and their implications, it also risks placing others, such as medical AI, in the same negative space.

The trope also relegates the very real threat of certain AI applications to the realm of fiction, causing some people to dismiss the reality. Further, it feeds the tension not just between man and machine, but also between supporters and detractors of the advance of AI. Reactions to this trope and the thinking it propagates make Hugo de Garis’s Artilect War seem increasingly possible.


For the above reasons, this trope may help check unrestrained progression of AI. Although still viewed by many as science fiction, potentially short-sighted and irresponsible AI creation has come increasingly under fire by notable scientists and technologists. And even though often flawed, this trope provokes discussion among the public in a way that encourages them to examine those concerns and increase their awareness in the advances and policies that may one day be their reality.

What do you think of the Homicidal Robot trope? Do you think the concern over homicidal AI is real?

Leave a Reply

Your email address will not be published. Required fields are marked *