Rebecca Ferguson and Chris Pratt Face Judgment in "Mercy"

By Laurence Lerman / New York City

Chris Pratt and Kali Reis are law enforcers of the future in Mercy

SCREEN TIME

Artificial intelligence has always made for a good movie villain, a seductive ally, or—most unsettling of all—a mirror. The new sci-fi thriller Mercy, which led the national weekend box-office with $11.6 million upon opening on January 23, arrives squarely in that tradition, offering a high-gloss, big-budget spin on anxieties that cinema has been practicing for nearly a century. Starring Chris Pratt as a detective of the near-future on trial for murdering his wife and Rebecca Ferguson as the advanced AI judge presiding over his fate, the film places its protagonist inside a 90-minute procedural countdown: prove your innocence or be condemned by a system you once championed.

It’s a sleek premise, built less for philosophical depth than for propulsive tension. But Mercy also lands at a moment when the cinematic language of artificial intelligence is beginning to feel uncomfortably close to home.

The film’s hook is elegantly cruel. Pratt’s detective,  Chris Raven, helped usher in a justice system run by artificial intelligence—one designed to be faster, fairer, and free of human bias. Now, accused of a deeply human crime, Raven finds himself pleading his case before an entity that cannot be swayed by charisma, history, or emotional appeal. The irony is baked in: the very tool meant to improve society may be incapable of recognizing the messy, contradictory truths that define it.

Alicia Vikander is an AI-infused humanoid robot in Ex Machina

Ferguson’s AI judge Maddox isn’t framed as overtly malicious; instead, she represents a system doing precisely what it was built to do. That moral ambiguity places Mercy in the lineage of AI-themed new millennium films like Minority Report, Ex Machina, I, Robot, and The Matrix—stories that ask not whether artificial intelligence can think or feel, but whether its logic ultimately aligns with human survival.

Neo gains the ability to perceive the code in The Matrix

Directed by Kazakh-Russian filmmaker (and tech entrepreneur!) Timur Bekmambetov, Mercy isn’t all that interested in metaphysics or ethics seminars. Rather, it’s focused on  genre pleasures: ticking clocks, courtroom faceoffs, the creeping dread that the machine is always one step ahead, and lots of high-tech action. This has long been Hollywood’s approach to AI—less a subject for sustained inquiry than a narrative accelerant. Artificial intelligence, in movies, is a storytelling device first and a philosophical question second.

That has been true since the beginning. Nearly a hundred years ago, Fritz Lang’s 1927 Metropolis introduced audiences to one of cinema’s earliest artificial beings: a gleaming mechanical double named Maria who’s been designed to manipulate the masses.

Maria, a robot, is energized to preach to the workers in Metropolis

Stanley Kubrick’s 1968 2001: A Space Odyssey offered HAL 9000, a calm, rational intelligence whose murderous turn felt chilling precisely because it was logical. The Terminator and its offspring transformed techanxiety into muscle and metal, while imagining a future in which machines don’t merely defeat humanity but render it obsolete, harvesting human bodies while trapping minds inside a simulation.

Across decades and genres, a pattern emerged. Artificial intelligence in films almost always produces an “us versus them” scenario. The machines evolve faster than we can control them, extend their reach beyond our comprehension, and inevitably threaten human existence—whether intentionally or as a byproduct of their programming. Even sympathetic portrayals tend to circle back to imbalance. In writer/director Spike Jonez’s sci-fi-spiked psychological romantic drama Her, the AI doesn’t destroy humanity; it simply outgrows it, leaving people behind with their feelings and their grief.

Astronaut David Bowman takes on the HAL 9000 computer in 2001: A Space Odyssey

Mercy follows that trajectory in a more grounded register. The threat isn’t global extinction or simulated reality (not yet at least)—it’s institutional dominance. An AI entrusted with judgment becomes dangerous not because it hates humans, but because it may not fully understand them. The film’s central fear is procedural rather than apocalyptic: what happens when justice is optimized to the point that it loses mercy? In that sense, the title feels less ironic than prophetic.

What’s striking, then, is that the films doing the most serious work around artificial intelligence right now aren’t thrillers or sci-fi spectacles at all, but documentaries. As AI moves from the speculative future to daily infrastructure—reshaping labor, authorship, power, and decision-making—narrative cinema keeps it at a safe remove, transforming the technology into villains, arbiters, or sleek existential puzzles.

Documentary filmmakers, meanwhile, grapple with messier present-tense questions. Who controls these systems, who profits from them, who is displaced, and what happens when opaque algorithms begin making human-scale decisions? Upcoming docs like Deepfaking Sam Altman, which follows filmmaker Adam Bhala Lough as he sets out to better understand AI technology and its creators, and The AI Doc: Or How I Became an Apocaloptimist, engage directly with the people, institutions, and incentives driving the AI boom. It’s not that fiction lacks imagination; it’s that documentaries are where imagination finally gives way to consequence.

Deepfaking Sam Altman humorously questions our relationship with emerging technology

Cinema has always embraced technological innovation, both behind the scenes and on screen. New tools promise greater realism, more spectacle, and more immersive experiences. Artificial intelligence fits neatly into that tradition. Mercy (arrives at a moment when AI is no longer an abstraction. It’s present in everyday life—in writing, art, surveillance, and systems that increasingly shape human outcomes. That proximity gives even familiar tropes a sharper edge.

The irony is that Mercy doesn’t need to be especially cerebral to resonate. Its power comes from recognition. We’ve seen this story before, in shinier or stranger forms, but the underlying question remains unchanged: when we build systems to be better than us, do we also build systems that can turn against us? Hollywood has been asking that question for a century, and audiences keep showing up for the answer—perhaps because, deep down, we suspect we already know it.

---

Laurence Lerman is a film journalis and a former editor of Video Business--Variety's digital media trade publication. Over the course of his four-decade career, he has conducted one-on-one interviews with just about every major filmmaker working today, from Martin Scorsese, Quentin Tarantino and Clint Eastwood to Kathryn Bigelow, Bernardo Bertolucci, and Werner Herzog. Most recently, he is the co-founder and editor-in-chief of the online review site DiscDish.com, the founder and curator of FilmShul.com, a multi-part presentation on the history of Hollywood and Jewish America, and a commentator on various 4K UHD and Blu-ray home entertainment releases.