A clip racking up more than 1.7 million views on Facebook has reignited a familiar online outrage cycle. The problem: the people in it do not exist.

The Claim

The video purports to show a street-style interview on Brighton beach in the UK. Two women, described as Muslim and wearing niqabs, appear to tell an interviewer that there are “too many dogs” in the UK and that dogs should be banned from public beaches. Commenters flooded the post with anti-immigrant and anti-Muslim remarks, and the clip was reshared across Facebook, TikTok, and X under captions framing it as proof that British culture is “under threat.”

Why It Is False

The video was not filmed. It was generated by an artificial intelligence video tool. Independent fact-checkers at Full Fact in the UK and AAP FactCheck in Australia examined the footage and confirmed it is synthetic. The original post came from a Facebook page called “Inside Australia,” which has a documented history of publishing AI-generated content dressed up as real street interviews.

Importantly, the interview never happened. No broadcaster filmed it, no journalist conducted it, and no real people were involved.

The Evidence

A few tells in the clip make the fabrication clear once you know where to look:

• A hidden digital watermark reading “Made with Google AI” is embedded in the footage and surfaces through reverse image search — a standard marker left by Google’s Veo video model.
• A dog visible earlier in the video inexplicably morphs into a handbag in a later frame — a classic sign of generative-AI temporal inconsistency, where objects shift because the model is not tracking physical reality.
• Facial features on both women subtly shift across seconds, and background beachgoers blur and rearrange in ways real footage does not.
• No UK news outlet reported the “interview.” There is no source clip, no outtake, and no reporter byline — a red flag for any piece of supposedly newsworthy vox-pop footage.
• The originating account, “Inside Australia,” has been repeatedly flagged by AAP for posting AI content as if it were real, and most of its other videos carry the same synthetic fingerprints.

Why It Spread

The video exploits a real formula: short, raw-looking footage that appears to catch ordinary people saying something provocative. That format bypasses the skepticism people usually reserve for polished news clips. When the content also maps onto existing political grievances — in this case, immigration and religion — algorithmic reach and emotional reaction do the rest.

How to Spot Clips Like This Yourself

1. Reverse image search a still frame. Tools like Google Lens and TinEye often surface AI watermarks or the original generated source.
2. Watch for morphing objects, inconsistent hands and fingers, warping text on signs, and backgrounds that shift between frames.
3. Check whether any real news organization reported the interview. Genuine street interviews almost always have a traceable source — a broadcaster, a reporter, a date, a location credit.
4. Search the account that posted it. A history of suspiciously dramatic “caught on camera” content is a strong signal the feed is engineered for engagement, not journalism.
5. Pause before sharing. Outrage is the payload synthetic media is designed to deliver.

Conclusion

AI-generated video is now cheap, fast, and convincing enough to fool millions of viewers in a single weekend. The Brighton beach clip is not an edge case — it is an early example of what an entire category of political and cultural misinformation will look like in 2026 and beyond. The defense is the same one that has always worked: slow down, check the source, and do not let a stranger’s algorithm decide what you believe.