Dear Parents and Carers,
Welcome to the second feature of our online safety newsletter looking at the use of technology and how we can work together to keep our children safe.
The digital world is an ever-changing landscape that can improve or hinder the development of our younger generation and today’s article is one which has hit the major news channels due to the severity of the dangers it now provides. A sensitive subject that we feel parents need to be equipped for if the unthinkable was to happen. As a safety precaution we do suggest that you do not do an internet search for some of the content of this article and please make a personal judgement on how much information you wish to share during conversations.
What are deepfakes and how to talk about them with kids?
What is a deepfake?
Deep-fakes are created using deep learning algorithms (a form of machine learning) to create pictures, videos, or audios that generate fake events – this is also where the name comes from. They are used to duplicate someone’s voice and/or facial features and paste them onto an existing recording or a photograph of someone else. The result depicts people in situations where they were never actually involved; this is also why deepfakes can be – and often are – used against people as a means of mocking and even bullying.
Simple versions of deepfakes can be easily made using mobile apps, such as FaceApp or FaceSwap, which children have access to within their own mobile phone. The world of Deepfakes is no longer a complex process often requiring advanced skills or serious technical equipment.
The uses of deepfakes
Deepfakes were previously used to discredit people and harm their careers, especially celebrities and high-profile people. Although reported, the dangers to children are now becoming more apparent.
Apps like FaceApp and FaceSwap allow users to make deepfakes without the need for specialist skills or equipment. These synthesised videos have now found their way into schools and become a vehicle for bullying. Imagine, for example, putting the face of a shy girl into a music video of scantily dressed women dancing provocatively. While the creators of such videos may view their actions as an innocent source of entertainment, the created deepfakes can cause the targeted child to experience shame. The unpleasant experience can easily damage their relationship with school.
Most recently the BBC have reported “A sleepy town in southern Spain is in shock after it emerged that AI-generated naked images of young local girls had been circulating on social media without their knowledge.” The pictures were created using photos of the targeted girls fully clothed, many of them taken from their own social media accounts.
These were then processed by an application that generates an imagined image of the person without clothes on.
So far, more than 20 girls, aged between 11 and 17, have come forward as victims of the app's use.”
(BBC News)
This is now a pertinent time to talk with children about deepfakes and how to explore preventative measures.
How to talk about deepfakes with children
1. Watch some of the deepfake videos together and start a conversation.
- Discuss why deepfake videos exist and what they can be used for. Share with them whether you enjoy the videos and why.
- Talk about responsibility and consent – explain why we should only use someone’s face or voice after getting their approval.
- Create a space of safe sharing, and tell your child that if they ever encounter the use of deepfakes for bullying, they can let you know.
- Make sure that children know whom to contact in case of need – their family and specified authorities in their school.
2. Try to spot the differences between deepfake and real videos together.
- Look for unusual movements, such as unnatural blinking. Deepfake videos tend to have issues replicating the subtler physical attributes:
- The audio may not fully correspond with the direction of the person’s lips, and there may be glitching along the lines of the person’s face or close to the hairline. There are also common discrepancies in the lighting – the face can be lighter/darker than the body, or there may be an unusual shadow, or each of the eyes may reflect a different image.
3. Talk to children about being online.
- Remind them that what they share online may be accessible for others to create a potentially harmful deepfake video of them.
- Look at what your children are sharing online. In addition to looking at what your children share, talk to them about what social media or apps they use.
- Look at what privacy settings that they are using to reduce access to their images.
4. Explore with your children and find preventative measures.
- Get to know the world they exist with online. Respect your child’s privacy, but perhaps ask them to scroll through their social media with you to see what type of content they consume.
- If they use apps that enable them to create deepfakes, such as FaceApp or FaceSwap, try them together and remind them that it shouldn’t be used to mock other people.
As a school, we are committed to working with parents to keep children safe online. If you have any questions relating to staying safe online, please contact Helen Kerr, Deputy Head and Designated Safeguarding Lead or John Roberts, Director of Learning for Computing and Business and strategic lead for the use of ICT in school.