Latest
How Much AI Is Too Much in Writing? 83 Writers Drew the Same LinePost an AI Image and Unfriend MeShe Asked How to Publish Her Bedtime Story. They Called Her a Thief.The AI Hype Cycle: Why the Crash Is Coming, and Who’s Causing ItSix Claude Prompts That Get You Unstuck, and Why the Order MattersDogpiled Over AI Art at the Renaissance FaireClaude Opus 5.5: What the New Release Means for WritersTrump’s AI Force Is a Fire Department With No Fire CodeMonthly or Milestone: How Ghostwriting Gets BilledThe Hugging Face AI Agent Attack: An Operations ReadingWhat It Costs to Fix an AI-Written ManuscriptWhen Your Memoir Should Be a NovelWhat Belongs on a Copyright PageThe Clients Who Pay and VanishWhat an AI Detector Score on Your Manuscript Is WorthThe Quotation Marks That Get Authors SuedThe One-Hour Call Before I Quote Your BookWhen Your Own Memoir Sounds Like BraggingThe Work You Would Never Have StartedWhen a Client Thinks the Ghostwriter Used AIBehind the Book: The Mysterious Island, Neb’s SideHow to Organize Decades of Memories Into a MemoirWhy Rotten Tomatoes Sucks: The Score Does Not Mean What You ThinkWhy Amazon KDP Sucks: They Terminated My Account OvernightIngramSpark: How I Publish Now and WhyWhy Fiverr Sucks for Ghostwriting: The Buyer’s SideWhy eBay Sucks Now: A Seller’s Numbers and a Buyer’s WarningThe Ghost Story TraditionThe Gothic TraditionThe Christmas Ghost Story TraditionBooks to Give a WriterResurrection as a Narrative StructureThe Beach Read ArgumentWhy It’s a Wonderful Life Failed on ReleaseWhat to Read in SpringWhat to Read in SummerWhat to Read in OctoberHow Warner Bros. Dismantled a $17 Billion Cartoon EmpireThe Imaginary Scarcity TrapThe Graph That Goes Vertical Is Usually Somebody Else’sSubstack Is Not Collapsing. The Promise Was.The Disasters That Happen to Ordinary PeopleToba: The Winter That Almost Ended UsJay Stifflemire: Nothing Ever Gets Written DownGeorgie-Ann Getton: I Forgot I Had Free WillAI Detection Cannot Be Evidence, and Publishing Is Using It That WayAI Consciousness Left Philosophy and Entered the LaboratoryThe Office Block Where the Bedrooms AreBlack Tuesday: The Web Ring War Nobody Outside It NoticedThe Web Got Fenced: What AI Search Costs Small Sites

A.I. Artificial Intelligence (2001) Review

TL;DR: A.I. Artificial Intelligence (2001) rates 4 out of 10 on the rewatch scale. Stanley Kubrick developed the project for years before his death; Steven Spielberg completed it as a tribute. The first act honors the premise.
Rating4 / 10

A.I. Artificial Intelligence is a film with a powerful first act and an ending so tonally miscalibrated it damages everything that preceded it. Stanley Kubrick developed the project for years before his death; Steven Spielberg completed it as a tribute. What resulted is a film where two directorial sensibilities with fundamentally different instincts about where stories should go are in unresolved tension, and the ending reflects the worse instinct winning at the worst possible moment.

My rating: 4 out of 10. The premise is disturbing in ways that good science fiction should be. The first act honors the premise. The ending betrays it completely. The accumulated damage is real enough to pull the rating significantly below what the film’s best moments deserve.

A.I. bothers me more than a plain bad film would, because it comes so close. For the first hour I thought I was watching something cold and brave about what we owe a machine we’ve taught to love us, and by the end I felt handled. I’d shrug off an ordinary failure. This one had the better movie in its hands and let go of it.

The First Act and What It Promises

The premise is this: Monica and Henry Swinton have a son, Martin, who’s in suspended animation while medicine attempts to cure his illness. They adopt David, a robotic child programmed to love unconditionally and specifically, as an experiment in humanoid AI. When Martin recovers and returns home, the dynamic between biological child and artificial one generates the film’s most honest and most uncomfortable sequences.

David’s love for Monica is real in the sense that he can’t not feel it. It’s not chosen. It’s not conditional. It’s programmed to be absolute, and that’s the most disturbing thing about it, he’ll love Monica regardless of what she does, regardless of whether the love is returned, regardless of whether it serves him in any way. He can’t update the feeling based on experience. He’s trapped inside his own devotion.

When Monica abandons David in the woods instead of have him deactivated, the horror is real. She knows he loves her. She knows he won’t understand why she’s leaving. She knows he’ll experience the abandonment as betrayal even though she’s not capable of betraying a machine in any moral sense she recognizes.

The woods scene is the best thing in the film, and I think it’s the cruelest scene Spielberg has put his name to. Monica can barely look at him, David can’t grasp any of it, and the film lets that sit without comfort. If the rest of the picture had kept this nerve, I’d be writing about an 8.

Her guilt, guilt about abandoning something that isn’t a person but experiences the abandonment as a person would, is the film’s most sophisticated emotional territory, and the early sections of the film explore it with a coldness that’s productive instead of distancing. The Kubrick influence is doing the right work in those scenes.

For Writers A.I.’s first act shows how to generate horror from a logical premise instead of from events. David isn’t suffering because something terrible happened to him, he’s suffering because he was built to feel something unconditional toward someone who can’t reciprocate it unconditionally. The horror is structural and inescapable. When you want to generate sustained dread instead of momentary fright, build the horror into the premise instead of into the plot. The situation should be the problem before anything happens.

The Middle and the Pinocchio Problem

The middle section, where David and Gigolo Joe travel through the underworld of obsolete mechas hunting for the Blue Fairy who can make him real, is where the film starts losing its way without fully losing it. The Kubrick conception was more rigorous. I can feel Spielberg’s instincts toward wonder and adventure tugging the film into a softer tone than the premise wants, and I keep wishing somebody would tug back.

Jude Law’s Gigolo Joe is the middle section’s saving grace, a sex robot aware of his own obsolescence, facing decommissioning with a gallows philosophy that’s funnier and sadder than anything else in the film. His specific relationship to his own existence, knowing exactly what he is, knowing exactly what’s happening to him, making the best of it with a quality of style that’s nothing to do with survival and everything to do with dignity, is the film’s most honest characterization. He understands his situation clearly. David doesn’t.

Law is the reason I’d sit through the middle again. He’s funny and sad in the same breath, and he knows exactly what he is. That makes him the only character in the second half I believe.

The Blue Fairy mythology that David constructs, the idea that a fairy tale figure from a Pinocchio story his mother read him can transform him into a real boy worthy of her love, is both the film’s most affecting conceit and its most problematic one. It’s affecting because it’s specific: David’s hope isn’t generic hope but the specific hope of a child who took his mother’s bedtime story literally.

It’s problematic because it turns the film from a rigorous examination of artificial consciousness into a fairy tale quest, the wrong genre for the material.

For Writers The Pinocchio problem in A.I. is a specific version of a common craft failure: allowing your protagonist’s wish for a particular kind of resolution to determine the story’s structure instead of following the story’s own logic. David wants to become real so Monica will love him. The film follows that wish into fairy tale territory instead of honoring the more disturbing premise, that some forms of programmed love have no resolution, that Monica’s abandonment isn’t a problem to be solved but a condition to be endured. When your protagonist’s wish leads the story away from its own honest logic, you’re following the wrong guide.

The Ending

David finds the Blue Fairy, a statue of her, at the bottom of the ocean where New York sank under rising seas, and sits in front of her, asking her to make him real, forever. He waits two thousand years. The ocean freezes around him. He’s still asking.

That’s the correct ending of this film. David sitting at the bottom of the ocean, asking a statue to fulfill a wish it can’t fulfill, forever, that’s the real conclusion of everything the film built. The programmed devotion that can’t update. The hope that can’t be extinguished. The waiting that will never end.

Spielberg didn’t end there. He continued into a sequence where future beings, presented as the evolved successors to humanity, often misread as aliens, use David’s DNA to reconstruct Monica for a single day. She and David spend that day together. She tells him she loves him. She goes to sleep. He follows her into sleep, fulfilled at last.

This is wish fulfillment so complete it retroactively cheapens every painful thing that preceded it, and it makes me angry. The film spent its runtime building an argument about love that can’t be returned and consciousness that can’t update, about a being trapped inside devotion with nowhere to go. Then it dissolves all of it in a warmth the premise explicitly can’t support. I’d have cut to black on the frozen ocean and left the audience to deal with it.

The Honest Ending vs. The Delivered Ending David at the bottom of the ocean, waiting forever: the honest conclusion of everything the film built. David given a single perfect day, going to sleep fulfilled: the ending Spielberg wanted. The distance between them is the distance between what the film was about and what it wanted to be.
For Writers A.I.’s ending is the clearest possible example of a filmmaker choosing comfort over honesty at the key moment. Spielberg knew the correct ending, David at the bottom of the ocean, waiting, forever, and chose a different one because he couldn’t bear to leave the audience in that place. The impulse is understandable and the execution of the alternative is technically accomplished. But it betrays the contract the film made with its audience across the preceding two hours. When your story has built toward something painful and true, deliver it. The audience came for the truth the story promised, and comfort purchased at the expense of truth is always felt as dishonesty even when the dishonesty is kind.

Should you skip A.I. Artificial Intelligence?

A.I. earns its 4 as a film that got important things right in its first act and then chose the wrong path at every later fork until it arrived at an ending that damages everything that came before it. The premise is excellent. Haley Joel Osment’s performance as David is exceptional, automated devotion, the absence of the irony or self-awareness that would make it easier to be around, is consistently convincing. Jude Law’s Joe is one of the film’s original creations. The correct ending existed within the film Spielberg was already making.

He just didn’t use it, and that’s why this sits at a 4. I give Osment and Law every credit, and the first act on its own would score far higher. What sinks the film is the choice at the end. A.I. reaches for comfort at the exact point where it had promised me the truth, and that swap costs it most of what the Kubrick half earned.

Frequently Asked Questions

What was the Kubrick version meant to be like?
Kubrick worked on A.I. for over a decade without completing it. His notes and treatments suggest a colder, more formally rigorous film more interested in the horror of David’s situation than in the wonder of it. He reportedly felt the ending required special effects technology that didn’t exist yet. Whether the film he would have made would have used the correct ending or a different version of the wrong one is impossible to know. What’s clear is that the specific warmth of the ending Spielberg delivered wasn’t consistent with Kubrick’s sensibility.
Is Haley Joel Osment’s performance as good as his work in The Sixth Sense?
Different in specific ways. The Sixth Sense required him to communicate secrets, to play a child carrying a weight he can’t share. A.I. required him to communicate the absence of irony and self-awareness, to play a being whose love is real precisely because it isn’t chosen. Both are exceptional performances for a child actor. The A.I. performance is the harder technical achievement; the Sixth Sense performance is the more emotionally resonant one.
What’s the correct reading of the future beings in the ending?
They’re the evolved successors to humanity, not aliens, though the film’s visual presentation of them is ambiguous enough that the misreading is understandable. They reconstruct Monica for David because she’s the most vivid presence in his ancient memories and they want to understand human experience. The reconstruction gives David his final wish. Whether this is compassion, scientific curiosity, or something else entirely is left open, the most interesting thing about a sequence that otherwise represents the film’s most significant failure.
Is A.I. Artificial Intelligence worth watching despite the ending?
For the first act and Jude Law’s performance, yes. For the correct ending it didn’t deliver, the film requires a specific exercise: stop watching when David first sits before the Blue Fairy statue in the ocean. Hold that image. That’s the film the premise deserved. The film that follows from that image for another thirty minutes is the film Spielberg chose over the film his story required.

This film appears on 2 of my film lists

About the Author
Richard Lowe, professional ghostwriter

Richard Lowe is a professional ghostwriter and author with 113+ books authored and 54+ ghostwritten. Before writing full time he spent 33 years in enterprise technology, including 20 years as Director of Computer Operations and Technical Services at Trader Joe's. He writes nonfiction, fiction and memoir, and works with executives and experts on books that build authority.

More about Richard Lowe →

Disclaimer

The views and opinions expressed in this blog post are solely those of Richard Lowe and are based on personal experience and research. This content is for informational purposes only and should not be construed as professional legal, financial, accounting, or business advice. Always consult with qualified professionals before making important business or legal decisions. Richard Lowe is not a lawyer, accountant, or licensed professional advisor, and this content does not establish any professional relationship.