Nikon contest AI photo row: Nikon strips its Small World in Motion winner after scientists spot signs of AI in the cilia video

The Nikon contest AI controversy wasn't about a photo. It was about a microscope video, the first-place winner of Nikon's 2026 Small World in Motion competition. On October 9, 2026, Nikon said the video "did not comply with the competition rules regarding generative AI" and adjusted the rankings, after scientists pointed out structures that appeared and vanished in ways real cells don't. It's the latest test of whether contests can tell AI-generated imagery from the real thing, a question that runs from science imaging to election campaigns, as our story on Netanyahu's AI image shows.

This story is based on Nikon Small World's October 9, 2026 statement and its competition rules, and reporting by ABC News (Australia), TechSpot, Digital Camera World, PetaPixel, IFLScience and CBS News, checked October 27, 2026.

The short version

• Which contest: Nikon Small World in Motion, the video side of Nikon's long-running Small World microscopy competition. This was its 16th edition.
• Which entry: Dr. Ning Xu's video of beating cilia in the airway of a child with primary ciliary dyskinesia (PCD), a rare genetic disorder, captured at 100x magnification.
• What scientists saw: structures that "pop in and out of existence," cilia that "appear from nowhere," and nuclei that didn't behave like nuclei.
• Nikon's decision (October 9): the video broke the generative-AI rules; Nguyen Nam Nhat of Vietnam, originally second, is now first.
• Xu's position: AI was used only to process the images, not to generate the cells or their motion, and he says he broke no rules.

What was the Nikon Small World in Motion video?

Xu's entry, announced as the winner in mid-September 2026, showed abnormal beating of cilia, the hair-like structures that clear mucus from airways, in a patient with PCD. Digital Camera World's September 21 winners story quoted him: "To understand this disease, you need to watch the cilia through dozens of beat cycles, not just a still image." It reported that he shaped light with a digital micromirror to keep the living sample unaltered and filmed it with diffractive super-resolution microscopy at 100x magnification. Xu was at Tsinghua University in China.

What did scientists spot?

Microscopists raised doubts within days, mostly in comments on a Nikon LinkedIn post that was later deleted, ABC News reported. Their main points, as reported:

• Impossible behavior. Edward Phelps, an associate professor at the University of Florida, said purple structures "pop in and out of existence" and the cilia "appear from nowhere," according to PetaPixel and TechSpot. Others said the nuclei didn't behave like nuclei and the cilia didn't match their known size. Phelps told ABC that "it's unacceptable to artificially render any part of a scientific image."
• A watermark. Ian Donovan, an MD/PhD student at UT Southwestern Medical Center, said that running the video through Google's Gemini detected an invisible SynthID watermark, Google's invisible marker for AI-made content, PetaPixel reported. That would indicate at least part of the video passed through a Google AI model. 9to5AI hasn't seen his check.
• Clinical experts. Robert Hirst of the University of Leicester, who has diagnosed PCD for 20 years, told the BBC he "knew immediately that the video was fabricated," according to Digital Camera World.

An open letter from clinicians and researchers who study PCD asked organizers to state clearly that the imagery contains AI-generated content, warning of a "potentially misleading portrayal of PCD." IFLScience reported it had 463 signatures from 36 countries and territories, including three former Small World judges.

On October 1, Nikon updated its LinkedIn post to say it was "carefully re-reviewing the information provided during the initial vetting," PetaPixel reported.

What did Nikon say?

Nikon Small World posted its statement on October 9, 2026. "After thoroughly re-evaluating the video and supporting materials and consulting with members of our judging panel, Nikon has determined that the winning video did not comply with the competition rules regarding generative AI," it said. It added that the decision "is based solely on the submitted video's eligibility" and "should not be interpreted as a judgment of the entrant's professional reputation, scientific contributions, or intent." The episode, it said, "has shown the need for Nikon Small World and Nikon Small World in Motion to revisit rules and evaluation procedures for future competition entries."

Nikon didn't say which part of the video broke the rules. Its current Small World in Motion rules page says: "AI generated videos are not permitted and Sponsor holds the right to request the original video for verification," and "Submissions generated by AI, script, macro or other automated means are void." That page now carries the entry period for the next competition, so its wording may differ from the 2026 rules.

Last verified: October 27, 2026

What does the entrant say?

Xu, a researcher rather than a photographer, says he didn't use AI to make the video. In a LinkedIn post since deleted, he wrote that AI was used to "distinguish and visualize features in the reconstructed grayscale images from the super-resolution optical imaging," TechSpot reported, and that it didn't generate the video, the cilia or their movement. Nature, as relayed by PetaPixel, quoted him: "It was applied afterwards to the reconstructed grayscale data to distinguish and color structures with similar morphology." He told ABC he rendered regions below the cilia to "improve presentation," not to "provide an anatomical reconstruction." He maintains he broke no rules, and says a fuller description of his optical system is being prepared for publication, according to Digital Camera World. Critics said sharing the raw grayscale footage could have settled the question, TechSpot reported.

Has this happened before?

Yes, in photography. In 2023, Boris Eldagsen won in the creative open category of the Sony World Photography Awards with "Pseudomnesia | The Electrician," then refused the prize, saying "AI is not photography," CBS News reported. The World Photography Organisation said he had confirmed the "co-creation" of the image using AI and removed him from the competition. The difference here is that Nikon's contest is about scientific imaging, where an AI-rendered detail can misstate biology, not just style. For which image models lead the public rankings, see our guide to the best AI image generators; for how watermarking schemes like SynthID fit into rules on labeling AI content, see our story on ChatGPT's EU text watermark.

What's confirmed, reported and uncertain

• Confirmed (Nikon): the October 9 statement, the generative-AI finding, the adjusted rankings, the $3,000 first prize and the rules' AI wording.
• Reported: the scientists' observations, the SynthID detection (PetaPixel), Hirst's comments (BBC via Digital Camera World), the 463-signature letter (IFLScience) and Xu's explanations.
• Uncertain: exactly which part of the video Nikon found AI-generated, and whether Xu's raw footage or forthcoming paper will be released.