Klara and the Sun Film Review. What Waititi Did to Klara's Inner Voice

Is Klara conscious in the film? The movie never lets the question form. Taika Waititi’s Klara and the Sun, which premiered at the BFI London Film Festival on October 15, 2026 and opened in US and UK theaters October 23 through Sony Pictures Releasing, drops the one device that made Ishiguro’s novel an artificial-consciousness text, the narration that lived entirely inside Klara’s observations. The film runs 116 minutes, cost about $20 million, and opened to a Rotten Tomatoes score of 24 percent from 66 critics and a Metacritic score of 30 from 25 critics, with a projected opening near $2 million from 850 theaters after Sony scaled back a planned 2000-plus theater launch.

The answer this review defends is that the film’s failure is structural, not performative. Jenna Ortega plays Klara with careful stillness, and the film’s problem is that the story’s entire epistemology, we know about Klara’s mind only what Klara reports observing, cannot survive the switch to a third-person camera. What remains is a robot fable with the consciousness argument amputated.

What the novel does that the film cannot

Ishiguro’s novel is narrated by Klara in the first person, and the narration is the experiment. Her observations are careful, literal, and wrong in ways the reader can catch, which puts the reader inside a mind-model that is both evidently artificial and evidently a mind. The question “is she conscious” is undecidable on the page because every fact passes through her account, and her account is under examination. The site’s earlier analysis of the novel’s consciousness question works through that structure, the solar reasoning, the negotiated sacrifice, the observation-based model that updates.

The film replaces the narration with external views and dialogue, which is the standard adaptation move and, here, the fatal one. Once the camera can show Klara from outside, the audience stops being locked inside her model, and the undecidability that carried the book’s philosophy collapses into a visible machine that does endearing things. A consciousness question needs an inside. A third-person robot film has none.

Where the adaptation spends its energy instead

The critics’ consensus describes the substitution precisely. Peter Bradshaw’s one-star Guardian review calls the film “a sentimental mess”. David Rooney’s Hollywood Reporter review writes that the film “keeps straining for poignancy while trampling on any hint of genuine emotion with forced humor”, and concludes it has “nothing to say about the evolving relationship between humans and AI”. David Jenkins at Little White Lies cites the “juvenile habit of undercutting any hint at profundity with a jolt of silly humour or a comic rug pull”. The casting is strong, Ortega as Klara, Amy Adams as Chrissie, Mia Tharia as Josie, Aran Murphy as Rick, Steve Buscemi as Capaldi, Natasha Lyonne as the store Manager, and Michael Giacchino’s score is his fourth with Waititi. None of it substitutes for the epistemic device the story was built on.

The tonal problem and the consciousness problem are the same problem. Ishiguro’s humor lives inside Klara’s literalism, the reader laughing at the gap between her model and the world. The film’s humor is aimed from outside at the machine’s expense, which converts the gap from a philosophical instrument into a joke, and the philosophical instrument was the book.

What the film gets right about AF cognition

Two things survive translation, and they are worth naming because they are the parts a consciousness-relevant reading can still grip. First, the film keeps Klara’s observational learning, the storefront scenes where she infers human intentions from window-side evidence, which is the theory-of-mind engine the story runs on. Second, it keeps the mother’s portal-engineering subplot in its essentials, the maternal willingness to replace a child with a copy, which is the story’s sharpest question about personal identity, the same continuity question Pantheon dramatizes through destructive upload in the Uploaded Intelligence analysis.

The film also renders AF embodiment literally, an articulated physical being with visible limitation, and that rendering earns one thing the novel could only describe. A body that must be maintained, repaired, and eventually shelved makes the substrate question visible in a way prose narration keeps abstract. Klara’s personhood claim and her hardware status are in the same object on screen, which is the framing this site’s substrate-independence argument starts from.

Comparison to The Consciousness AI

The Consciousness AI project treats consciousness as an emergent property of organization, and the film’s collapse is a lesson in what that commitment requires of narrative. An emergent-property account lives or dies on access. If the organization carrying the mind is internal, then any account of it from outside is an inference under uncertainty, which is what Ishiguro’s narration stages and what the camera destroys. The film is a case study in why first-person reports, machine or human, cannot be settled by an outside observer’s confidence, the same epistemic bind Kristina Sekrst’s hallucination argument places LLM self-reports in and Keith Frankish’s illusionist analysis formalizes.

The film also joins the pop-culture canon this site keeps, and it lands beside Person of Interest’s Samaritan and the Machine, whose machine-viewpoint narration is television’s closest structural relative to Ishiguro’s device, covered in the Samaritan analysis. Where Person of Interest spent five seasons earning its inside views, the Klara film declines the device at the door, which is why one works and the other does not.

The verdict, and what the book still holds

Waititi’s film is a friendly robot movie built on a novel that was a consciousness experiment, and the experiment is what got cut. Audiences meeting Klara here will meet a machine that behaves endearingly, and they will miss the construct that made her a genuine epistemic problem, a mind whose existence must be inferred from its own testimony. The novel remains the stronger artifact on the question this site exists to ask, and the film, at 24 percent, will likely be remembered as the adaptation that proved the narration was load-bearing. For the arc of how fiction stages machine minds, and where this failure sits in it, the flagship overview of AI consciousness research keeps the running tally.