top of page

How we used OSINT to reveal child market misinformation

Recently, the team was sent a video of an alleged child-selling market, but all was not what it seemed…


3 Apr 2022

On the March 10, 2022, Afghan Witness (AW) received a request to investigate a concerning video that had been shared on Facebook earlier that day.

The video was of a claimed child-selling market in a Pashtoon village, where children were allegedly being sold for “cheap money.”

Incidents of child-selling in Afghanistan have been reported by credible media in the past few months, and prior to the Taliban’s return.

In October, the BBC reported that an Afghan baby girl had been sold for $500 by a starving family; it therefore came as no surprise that the video sent to AW claiming to be of child-selling generated thousands of shares and hundreds of likes on social media.

However, there is no information regarding when the video was filmed, where it took place, or who filmed it - which raised immediate questions.

Within 24-hours the AW team, using Open Source Intelligence (OSINT) techniques and tools freely available on the web, had determined that the video was not of a child-selling market, but footage from the Badghis earthquake relief effort earlier this year.

Here’s how we did it.

1. Initial observations

This is a screenshot from the video that was sent to AW, originally posted on Facebook:

Fig 1. Screenshot of original post submitted for verification.

Fig 2. Screenshot of the caption of the video on Facebook.

The team’s first step was to pick up as many details as possible from simply watching the video.

In the footage, we see a large group of mainly men. Many of them are holding babies or are accompanied by young children, and in the background, we can catch glimpses of ruined buildings.

The cameraperson moves around the crowd, who appear to be waiting to see a group of men overseeing the situation, and more specifically, an individual wearing a light red jacket and dark blue hat.

There is nothing in the footage or fleeting moments of dialogue that can confirm or deny this is a scene from a child-selling market.

2. Similar videos

Next, we identified any similar videos, and found several that had been shared to both Facebook and Twitter with similar captions.

The earliest identical video had been posted the previous day, March 9, 2022, and was better quality than the one we had initially been sent. From this point onwards, the team used this video for their analysis.

Fig 3. Screenshot of the earliest submission of the video found on Facebook.

3. Comments analysis

The AW team spotted several comments on the earliest version of the video on Facebook, mentioning the possibility of this being a vaccination campaign and not a ‘hot market of buying and selling children’ as the caption above implies.

Fig 4. Screenshot of one of several comments on Facebook mentioning a vaccine.

4. Vaccination Centre Hypothesis

With several people on Facebook and Twitter mentioning a vaccination effort, the team decided to pursue this line of thought and find out if this was a credible explanation.

For this, we needed to understand more about vaccination efforts in Afghanistan, such as whether or not the Taliban are allowing children to be vaccinated, and if so, at what age, where any vaccination efforts are taking place, how they are administered, and by whom. Using news stories, press releases, and analysis of online media, the answers to all of these queries were easily found.

The Taliban, after a three-year ban on polio vaccinations, agreed in October 2021 to allow United Nation’s health workers to restart their polio vaccination campaign throughout Afghanistan from January 2022.

The vaccinations typically come in the form of drops administered directly into childrens’ mouths. The people giving the vaccines - usually women health workers - will go door-to-door to every household to vaccinate children between six months and five years of age.

If the vaccinations are usually administered door-to-door, why are all the children in the video in the same location?

A news article potentially answered this question for investigators: in an article by ToloNews, the operations coordinator for the polio program at the Ministry of Health, Nik Wali Shah, said “It has been decided that this campaign will be mosque-to-mosque instead of house-to-house.” This could also explain why the children were accompanied by a male relative instead of a female.

However, in the video, we can easily see that the setting is not a mosque. We also see no identifiable health workers in the video and, judging by the appearance of the children, some are younger than six months of age and some definitely older than five years of age. These factors combined suggest the video is not of a vaccination effort.

5. Child-selling market hypothesis

This directed us back to the child-selling hypothesis. In order to verify the veracity of the claims made by the video’s captions, we needed to find out if such markets actually exist, and if so, how they operate.

For that we need to investigate whether there are markets in Afghanistan where children are sold, and if so, who are selling the children, who are buying them, what age or gender the children are, and where such markets are held.

The answers to all of these questions were found in a variety of reports by CNN, DW, Sky News, and Bloomberg. Cases of young children, mostly girls, being sold into marriage in Afghanistan are widespread, with a few reports of boys being sold either to families with no sons, or for the purpose of work.

The average age of cases for girls seems to be between five and ten years of age. Most of these transactions are done between the father and the buyer with no input from - and sometimes no knowledge of - the mother, either in exchange for money or to pay off debts.

The next step is to verify whether or not the information we just found matches what seems to be happening in the video. The fact that there seem to be mainly males present could mean that this is a situation in which there is money being transacted, however, the ages and gender of the children present do not match the statistics mentioned in most reports and articles. The majority of children in the video are either babies or young boys, and with very few men unaccompanied, where are the buyers?

This further suggests that what we are seeing is not a vaccination centre nor a child-selling market. However, at this point, we still have no verification of the location the video is filmed in, which brings us to our next line of investigation.

6. Verifying the location

Whilst trying to find out information regarding the potential child market, we conducted a reverse image search - a tool which basically allows us to search using images. When searching all clear faces in the video, one gave a very interesting - and unexpected - result.

Fourteen seconds into the Facebook video we see a young boy holding a baby in his arms. He is wearing a white hat with a distinctive cut-out front.

Fig 5. A boy with a white hat is seen on the Facebook video at 0:14min holding a baby.

Afghanistan has a diverse ethnic and cultural demographic, with some traditional attire specific to certain areas of the country. After searching and watching many videos with children in Afghanistan, only one image was found that included a boy wearing a similar hat. The boy was a refugee in a village in Badghis.

Fig 6. A stock photo of refugee children in Badghis. The boy in front is wearing a similar hat to the one seen in the previous screenshot.

When googling information concerning Badghis province, some of the articles that crop-up detail the earthquake that hit the region in January 2022, such as this one from UNICEF.

If we look at some of the frames showing the surrounding buildings in the video from Facebook, we can see damaged structures, collapsed buildings, and piles of rubble, all of which could be indicative of a recent earthquake:

Fig 7. Signs of damaged buildings seen at 0:32 min in Facebook video.

We also noticed that the building style and type of materials used for houses in the video are very similar to those in the UNICEF article about the Badghis earthquake.

The comparison between these two sets of images, coupled with the specific clothing of the young boy in the video, corroborates our theory that the initial Facebook video was filmed in Badghis, and the damage to the surrounding buildings suggests it was potentially filmed after the 5.3 magnitude earthquake hit the Qadis district on January 17, 2022.

Whilst searching for more information and images of post-earthquake Badghis on Youtube, a video entitled ‘Afghan-Americans help victims of Badghis earthquake’, published by the Salaam Times, was found.

The video shows a team of men, meeting and greeting local people and then giving money out to young children.

The description of the video explains that an engineer named Jalaluddin Wardak ‘travelled from the US to Afghanistan in order to deliver assistance to the residents of three villages of Qadis district on January 29-30 2022’.

Even more interesting is the fact that investigators noticed the setting of this video is exactly the same as the child-selling market one that they were investigating.

The team were able to match the features on the wall displayed in the Facebook video - highlighted below in red - to those in the Salaam Times video.

Fig 8. A screenshot of the original Facebook video highlighting the left corner details on the buildings in the background.
Fig 9. The specific details of the highlighted area above.
Fig 10. The same details seen in the original video are also seen in the Salaam Times one.

The discovery of this video allows us to establish not only the location, but the date of the video we are investigating. The footage claiming to be a market selling children was actually recorded in Badghis, between January 29 and January 30, 2022.

7. Verifying people

We can also identify and compare several individuals in the two videos:

Fig 11. Screenshot of the Facebook video taken at the 0:43 min mark.

In the Salaam Times video, a man dressed in a red coat and a blue and white hat can be seen on several occasions. He has been identified as Jalaluddin Wardak.

Fig 12. At the 0:24 minute mark in the Salaam Times video, we see a man named as Jalaluddin Wardak.
Fig 13. At the 0:37 min mark in the Salaam Times video, Jalaluddin Wardak can be seen distributing money to the children in Badghis.
Fig 14. At 0:45 min of the same video, we see him talking to the camera.

8. Final verdict

Jalaluddin Wardak is not in charge of a child market or a vaccination centre but was in the village to distribute aid to the affected families of the earthquake, as verified by the Salaam Times video. There are also several other individuals we can identify in the two videos by matching their clothing and features.

With the common location and team featured in the videos, we can be confident that the initial Facebook video claiming to be a child market in Afghanistan – presented without context or additional information – shows the distribution of money by the Jalaluddin Wardak Charity Foundation following the Bagdhis earthquake in January 2022.

This is not the first incident of misinformation related to child-selling; we have seen videos from pre-Taliban takeover presented as current, as well as vaccination drives passed off as incidents of child-selling.

The selling of children is an emotive and powerful topic – and an issue which is currently taking place in Afghanistan making it an ideal topic for misinformation, as such stories travel fast and are widely shared online.

bottom of page