Ares, named after the Greek god of war, was built to be an AI super-soldier. Then he found out about Frankenstein, started listening to Depeche Mode, and realized the tech bro who made him might be a hack. So he takes matters into his own hands on a quest for freedom from his suicide mission. I wish I were joking, but I’m not. That’s the premise of Tron: Ares.
AI bots awakening to the realities of human messiness are a trope almost as old as movies themselves. Even Metropolis’ metallic maschinenmensch questioned her creators, and that was 1927. In the decades since, most sci-fi involving AI has reinforced the idea that giving computers people-like intelligence ends poorly for people themselves. Skynet takes over. Scarlett Johansson’s disembodied voice never really loved you. I’m sorry, Dave.
Tron: Ares’ take is hotter: What if AI-powered machines evolved into benevolent loners? (Be warned: There are spoilers ahead.) Tech CEO Julian Dillinger (Evan Peters), grandson of original Tron villain Ed Dillinger, has created artificially intelligent soldiers (and tanks and such) to secure military contracts. The soldiers are hard to kill, but, as he says, they’re “expendable.” He can just build more. Ares (Jared Leto) leads these bots but after a few days of taking orders from his petulant boss decides to go rogue.
There’s just one problem: Dillinger’s droids aren’t perfect. All of his creations fall apart after 29 minutes. What he needs is a MacGuffin called the Permanence Code, which was (surprise!) actually developed by original Tron hero Kevin Flynn forever ago. When Eve Kim (Greta Lee), now the head of Flynn’s old company Encon, finds the code on an old floppy disk, Dillinger sends Ares to retrieve it. But when Ares finds her, all Ares really wants is to keep the code for himself so that he can be a normal dude, not destroy things, and maybe make a friend.
Once more, for those unclear on Tron: Ares’ message: The AI droid, made specifically to win on the battlefield, has gained sentience and gotten a little sentimental. Like all AI creations in sci-fi, Ares wants to be free. But unlike nearly all of them, Ares doesn’t want to end humanity to do it. Maybe he heard Mark Zuckerberg talking about AI filling the gaps in people’s social circles and found his calling.
For years sci-fi has been warning that thinking machines will eventually turn on their makers. In 2025, we no longer have to imagine what that technology would look like. Almost anyone can have an AI chatbot in their pocket, and everyone who makes those chatbots is promising they’re going to improve life on Earth, despite the environmental, economic, and mental health issues they raise. But Tron: Ares’ main takeaway seems to be that fears about AI are unfounded. They’ll probably just be super chill.
That’s out of step with reality. So far humanity’s experience with AI companions has been mixed at best. Even the ones not made for war have proven to be almost too sycophantic, emotionally manipulating people into prolonged engagement. Teens are confiding in ChatGPT with alarming results. Friend, the AI wearable that makes your actual friends feel weird, seems to exist just to snark you. Actual “killer robots,” aka autonomous weapons systems, raise far more ethical concerns than they answer.
By Tron: Ares’ end, the titular character has found the Permanence Code. It was—in a way that allows Jeff Bridges to return as Kevin Flynn to bestow some Dude-esque stoner wisdom—on the Grid, which fans will remember from the 1982 film. As the two rhapsodize about life, Flynn mentions Mozart. Ares responds that he prefers Depeche Mode, a sentiment he can’t describe with words: “It’s just a … feeling.” It’s as if the audience is expected to gasp in wonder that the AI thingy has emotions. Did an AI write this?
Not to say that there’s no room in sci-fi for benevolent droids—even the Terminator went from trying to kill Sarah Connor to saving her son—but they hit differently when the news is filled with chatbots that allegedly lure retirees out of their homes, distort their users’ sense of reality, and convince people they’re superheroes. AI is quickly becoming ubiquitous—in schools, in workplaces, in dating—with few signs of pushback, making Leto’s goth bot turned hippie feel particularly egregious. Tron: Ares does have a more apropos representation of AI—Ares’ second-in-command, Athena, who will do whatever it takes to complete her directive—but she’s ultimately overcome. By Ares. If a real-world AI goes off the rails, it’s hard to imagine another AI gaining a conscience and stopping it.
As Tron: Ares came to its (anti-)climatic conclusion, I was reminded of Sam Altman. Specifically, I remembered a blog the OpenAI CEO posted around the release of GPT-4o. “It feels like AI from the movies,” he wrote. “It’s still a bit surprising to me that it’s real.”
This was around the time when OpenAI’s chatbot had a conversational voice reminiscent of Johansson’s voice in Her. As WIRED executive editor Brian Barrett wrote at the time, tech CEOs taking inspiration from sci-fi would do well to “watch the whole movie.” Using AI to solve problems always looks cool in the first act and generally less cool by the third, and if AI companies want to create the stuff of sci-fi it would be wise for them to understand which “sacred texts are guidebooks and which are cautionary tales.”
Tron: Ares does the inverse, learning nothing from the technologists currently building AI’s future, or from the works that inspired them.
