BBC Explains How It Verifies Viral News Videos (Full Transcript)

A step-by-step look at how journalists confirm where, when and why social media footage was filmed using satellite images, sun tools and reverse searches.
Download Transcript (DOCX)
Speakers
add Add new speaker

[00:00:00] Speaker 1: When a news story breaks, social media is often flooded with videos claiming to show what's happened. Some are genuine, others are misleading or even false. Our first priority is to work out which is which. When we can, the BBC will go directly to the scene where it happened to check. But this isn't always possible. Some areas are too dangerous, too remote, or do not allow journalists independent access. So, footage posted online might be the first or only evidence we have of what's happened. But how do we verify a video? We need to answer three things. Where it was filmed, when it was filmed, and why it was posted. Let's start with where. We look for clues in the video. These could be road and shop signs. We also listen for voices. What language are they speaking? Is it in a local dialect? We also compare what we see in the video with satellite imagery. Are the buildings the right shape? Do the trees, roads and terrain match? Satellite images can also help us confirm changes to the landscape, such as damage by attacks or ongoing conflict. Next, we ask when it was filmed. If it's outdoors, we can use the weather, shadows, the position of the sun to work out what date and time the video was filmed. We use online tools like SunCalc to calculate where in the sky the sun should be at any given time. Or weather websites, which keep records of sun, rain, cloud cover and wind direction. We also use a technique called reverse image searching. We take multiple screenshots from a video and check whether those have appeared online before. If we find them months earlier in a totally different context, well, then we know it's not what it claims to be. Sometimes it can lead us to the earliest copy online and the person who posted it. We often contact them. They may well have filmed it and they will have vital context. So finally, if we can, we ask why. Why was this video uploaded? Who posted it? What have they said about it? And what might their motives be? If we can answer where, when and why, well, we can be confident something can be considered verified.

ai AI Insights
Arow Summary
The speaker explains how the BBC verifies videos circulating on social media during breaking news. Verification focuses on determining where the video was filmed, when it was filmed, and why it was posted. Location is checked through visual and audio clues (signs, language/dialect) and by matching features to satellite imagery. Timing is assessed using weather conditions, shadows and sun position with tools like SunCalc and historical weather records. Reverse image searching of video screenshots helps detect recycled or miscontextualized footage and can lead to the earliest uploader, who may be contacted for context. Finally, understanding the uploader’s identity and motives helps assess reliability. Answering where, when, and why supports confidence that the footage is verified.
Arow Title
How the BBC Verifies Social Media Videos
Arow Keywords
video verification Remove
BBC Remove
social media Remove
breaking news Remove
geolocation Remove
satellite imagery Remove
dialect analysis Remove
SunCalc Remove
shadow analysis Remove
weather records Remove
reverse image search Remove
misinformation Remove
source verification Remove
context Remove
open-source intelligence Remove
Arow Key Takeaways
  • Verify footage by answering three questions: where, when, and why.
  • Geolocate videos using signs, terrain, building shapes, language/dialect, and satellite imagery.
  • Determine filming time using weather, shadows, and sun position with tools like SunCalc and historical weather data.
  • Use reverse image searches on multiple screenshots to detect reused footage and find the earliest upload.
  • Contact original posters when possible to obtain context and corroboration.
  • Assess the uploader’s identity, claims, and potential motives to understand why the video was shared.
Arow Sentiments
Neutral: The tone is explanatory and procedural, focusing on factual steps and tools used to verify videos without emotional language or advocacy.
Arow Enter your query
{{ secondsToHumanTime(time) }}
Back
Forward
{{ Math.round(speed * 100) / 100 }}x
{{ secondsToHumanTime(duration) }}
close
New speaker
Add speaker
close
Edit speaker
Save changes
close
Share Transcript