Warning to Parents: Children's Photos on the Internet Become Fuel for AI Crimes

Listen to the article. The audio text is automatically generated by a machine system.

Generative artificial intelligence can transform a short sentence written by the user into a highly realistic image within seconds, but this capability is increasingly being exploited by criminals to create images of sexual assaults on children, in a problem that is rapidly escalating.

According to the Guardian newspaper, regulatory bodies detected materials of this kind during the first six months of 2026 that exceeded what was recorded throughout the entire year of 2025, highlighting the risks of sharing images of children online, even when the original images are completely ordinary.

The Internet Watch Foundation (IWF), a British organization specialized in detecting and removing content related to child exploitation online, identified 6,310 illegal AI-generated images between January and June 2026.

In comparison to last year, when the organization detected just over 4,500 images, this represents an increase of about 40%.

The majority of the images targeted girls, and most of the children who appeared in them were between the ages of 7 and 13.

The danger is not limited to images created by criminals from scratch; real images published online can also be manipulated and turned into explicit content using artificial intelligence tools.

The Report Remove service, which helps remove intimate images from the internet, received 420 reports from children who said someone might have altered their photos using digital tools to make them look like explicit images.

Notably, this number has already surpassed the total reports received by the service throughout the year 2025, which amounted to 397 reports.

This reflects a troubling shift in the nature of the danger; criminals no longer need to obtain real explicit images, but can use ordinary images of children published online and modify them to produce abusive content.

Britain is already treating AI-generated images depicting child sexual abuse as a crime.

Modifying an AI model to enable it to produce this type of content, or sharing a model capable of producing it, is also against the law.

Developing a model specifically for this purpose could lead to imprisonment for up to five years.

However, the Internet monitoring organization believes that current laws need to be tightened further.

Hannah Suersky, the policy officer at the organization, stated that there is a need for mandatory AI laws that force companies to design safer models from the outset, rather than trying to address their misuse after they are released.

The technology companies bear the responsibility of developing tools that cannot be misused in this way.

There are currently no indications of a specific British law being proposed for artificial intelligence that addresses this issue directly, but Kanishka Narayan, the British Minister for Artificial Intelligence, stated that nothing is off the table, noting that any new proposal will be evaluated based on its ability to enhance the safety of Britons.

Both the National Crime Agency (NCA) and the Internet Watch Foundation advise families to stop posting pictures of children in general and openly online.

If posting pictures is necessary, it is better to switch social media accounts to private mode or limit the sharing of pictures to a small circle of trusted friends and relatives.

It may be difficult to resist the urge to post a picture of your child on their first day of school or to share a special family moment with friends, but the increasing prevalence of artificial intelligence tools is changing the nature of the risks associated with these images.

Spending a few minutes today reviewing your privacy settings and limiting who can see your children's photos can significantly reduce future risks to family privacy.