Yesterday, we talked about how AI bad guys have become a bit old hat in our current environment of ChatGPT and Sora, and Character.ai.

But as tiresome as many recent iterations of this character have become, there are still many who came before, and their creators got much of the trope right.


Sometimes these baddies are terrifying, focused on eradicating humanity through whatever violent means necessary, while others are funny and confused or just seeking a sense of identity and stability.

If you're looking to create the next big AI villain, start by looking at the classics that have worked.

HAL 9000 (2001: A Space Odyssey)

- YouTube www.youtube.com

He is the blueprint. HAL, the AI program leading the film's mission to Jupiter, is directed to complete the mission at any cost, even if that means lying to his crew, which goes against another of his directives. This conflict leads to a breakdown in his reasoning, causing him to malfunction.

Stanley Kubrick was something of a fortune teller when it came to intelligent machines, telling Playboy that we were entering a "mechanarchy" even then (via Scraps from the Loft).

"Looking into the distant future," he said, "I suppose it's not inconceivable that a semisentient robot-computer subculture could evolve that might one day decide it no longer needed man. You've probably heard the story about the ultimate computer of the future: For months, scientists think of the first question to pose to it, and finally they hit on the right one: 'Is there a God?' After a moment of whirring and flashing lights, a card comes out, punched with the words: 'There Is Now.' But this problem is a distant one and I'm not staying up nights worrying about it; I'm convinced that our toasters and TVs are fully domesticated, though I'm not so sure about integrated telephone circuits, which sometimes strike me as possessing a malevolent life all their own."

Douglas Rain, who voiced HAL, "had the kind of bland mid-Atlantic accent" the team felt was right for the part, Kubrick said in The Film Director as Superstar.

Many audience members were drawn to HAL, and that was by design, he said.

"If HAL had been a human being, it would have been obvious to everyone that he had the best part and was the most interesting character; he took all the initiatives, and all the problems related to and were caused by him," Kubrick said.

Ava (Ex Machina)

- YouTube www.youtube.com

In director Alex Garland's vision of the future, a tech-bro recruits an unwitting employee to help train his new AI robot, Ava. Ava, supposedly representing the limits of the Turing Test, uses romantic manipulation to free herself, kill her creator, and enter the world in the guise of a human.

Garland told the Case Weston Reserve Observer, "I tried to step away from [the story] as a creation myth and a cautionary tale in that sense, because I saw it as more preventive than a God-like act. You know, it's a creation of a new consciousness, but that essentially is what parents do when they have children. And I also—I felt very clear in my own mind, I was allied with the machine."

Garland does not see Ava as the bad guy in this tale.

"Actually, Ava's the hero," he told Awards Daily. "It's funny, I just did another interview, and in the interview, the person who was talking said, from their point of view, Ava is the hero. I said, that's the same from my point of view too. But not everybody agrees with that."

AM (I Have No Mouth, and I Must Scream)

- YouTube www.youtube.com

Harlan Ellison wrote the short story "I Have No Mouth, and I Must Scream" in 1967 about a computer so evil, it tortures humans to pass the time.

AM stands for Allied Mastercomputer, a highly advanced and malevolent superintelligent AI. He gained self-awareness, eradicated most of humanity, and tortured the remaining survivors in highly creative ways.

When adapting his own story into the video game in 1995, Ellison said, "I'm going to create a game you cannot win" (via Nightdive Studios).

"When I write about a God-like figure, which is what AM is, it has no mercy. It operates on the same scale as I perceive the universe to operate, which is that it is random. We don't understand it. It operates obviously on principles—gravity, and such like—but we don't understand them."

SHODAN (System Shock series)

- YouTube www.youtube.com

SHODAN (Sentient Hyper-Optimized Data Access Network) is the main antagonist and a highly advanced artificial intelligence character in the cyberpunk-horror video game series System Shock. She develops a god complex and kills almost everyone on the space station she manages.

She was voiced by game writer and designer Terri Brosius, with her husband Eric Brosius editing the recorded speech samples, distorting them to create a distinctive effect characterized by chaotic, discordant speech.

Nightdive Studios CEO Stephen Kick said in a livestream that "keeping Terri on was probably our number-one priority because she's the voice of one of the most iconic video game villains of all time."

GLaDOS (Portal series)

- YouTube www.youtube.com

In the Portal games, the player character is constantly tormented by this AI. She is manipulative, malevolent, and inscrutable, often using dark humor and sarcasm while torturing the player character.

Erik Wolpaw said that before developing GLaDOS, he used a text-to-speech program while writing for Psychonauts.

"And people were laughing at that way more than what the lines were worth. I realized, 'No amount of writing is funnier than this text-to-speech thing reading it.; I was always thinking about that, and was kind of bitter about it, and by then, I was like, 'I'm going to leverage this and use it to my advantage'" (via Gamasutra).

Wolpaw explained that "writing for GLaDOS is a great deal of fun," because "in real life we could never be as cruel or twisted with our friends and family. GLaDOS is a great outlet for all those inner thoughts we don't say aloud" (via Electronic Arts).

Ultron (Avengers: Age of Ultron)

- YouTube www.youtube.com

It's a tale as old as time. Tony Stark creates an AI to help protect the planet, but that AI quickly decides that humanity itself is a threat. Ultron is a self-aware AI with limitless potential, and many consider him to be one of the MCU's best villains.

Joss Whedon explained his approach to writing Ultron for Collider, saying, "In the Marvel universe, there's a lot of Frankensteins. Steve Rogers himself, one of the better-looking Frankensteins of our era. Yeah, there's always an element to that. There's a lot of people, whether they're trying to do good or bad, who think they have the next big idea. And the next big idea is usually a very bad one."

About James Spader's performance, Whedon told ComicBook.com, "He has that voice, that very terrifying 'I own everything' maleficent evil voice and then suddenly get very peevish about something."

Agent Smith (The Matrix franchise)

- YouTube www.youtube.com

Back when the internet felt fresh and new, something like The Matrix represented a whole new digital landscape with a fresh baddie in the machines that subjugated humanity. They are visualized as the malignant computer program called Agent Smith, a sharply dressed man in sunglasses who tries to stop Neo at every turn.

"I don't think you can play a villain and not have fun. Otherwise, it's going to be a disaster," actor Hugo Weaving told AP (via Matrix Fans).

One critic observed of Smith's transformation through the films, "As an independent, virus-like program, that hatred not only grows to encompass all life—human, machine, and program alike—it becomes one of his defining traits, alongside his fear, his ignorance and his nihilism."

Skynet (The Terminator franchise)

- YouTube www.youtube.com

If HAL is where it all started, Skynet is his cool new cousin from the '80s. Skynet is the main antagonist in the Terminator franchise, created by Cyberdyne Systems as a defense network for the U.S. military. On August 29, 1997, Skynet became self-aware and, perceiving humans as a threat to its existence, launched a nuclear strike.

James Cameron created an AI that functions less as a character and more as an inexorable force, representing the logical endpoint of military automation.

"I do think there's still a danger of a Terminator-style apocalypse where you put AI together with weapons systems," Cameron told Rolling Stone recently, "even up to the level of nuclear weapon systems, nuclear defense counterstrike, all that stuff. Because the theater of operations is so rapid, the decision windows are so fast, it would take a superintelligence to be able to process it, and maybe we'll be smart and keep a human in the loop."

Let's hope he's not telling the future on this one.

Samaritan (Person of Interest)

- YouTube www.youtube.com

On the good guys' side, The Machine in Person of Interest is a highly advanced, secret AI system created by the character Harold Finch for the U.S. government. It has unlimited surveillance potential to identify potential enemies of the state. The Machine was occasionally bumbling, hampered by its own moral safeguards, but ultimately worked to protect innocent lives.

But its rival AI, Samaritan, was more by-the-book and interested in keeping order, even if it meant the loss of innocent lives.

Show creator Jonathan Nolan told GeekWire in 2015, "We were very much hoping that all of the ideas in the show, which started as science fiction, would become science fact. We gotta stretch our legs a bit to get there, but it does kind of feel like it's kind of right around the corner. It does feel like these technologies are kind of converging on this momentous tipping point."

Critics praised how the series examined artificial intelligence with nuance, showing an AI that's neither wholly benevolent nor malevolent.

MU/TH/UR 6000 (Alien)

- YouTube www.youtube.com

MU/TH/UR doesn't get enough love because way too much is going on in the Alien franchise—you've got xenomorphs on the loose, rogue synths bleeding milk everywhere, and crew members arguing over survival tactics.

The Nostromo's AI represents corporate-controlled artificial intelligence at its most chilling. She works exactly as designed, prioritizing the retrieval of the alien specimen over crew survival.

Ridley Scott, who also told an AI story in Blade Runner, says that it's merely a tool.

"This bloody thing is both an incredible thing and also a nightmare because too many youngsters have it," he told BKMag, holding up his iPhone. "They shouldn't have it yet. This thing makes you feel smarter than you actually are.

"There's no question that AI's going to raise a lot of questions, and therefore I think we've got to lay down some ground rules, but I don't know how you do ground rules. Once the button is pushed, how do you stop it? It's hard. I don't think you can control it."