The biggest AI stories of the past seven days, explained through a family safety lens
This week’s AI news carried more than the usual headlines about tech companies and market share. Several major developments directly affect how children experience the internet, who or what they are talking to online, and the risks they face from AI-generated content. As a parent or guardian, here is what you need to understand.
1. Canada Has Just Introduced a Law to Ban Under-16s From Social Media
On 25 June 2026, the Government of Canada introduced Bill C-34, the Safe Social Media Act. The legislation includes a minimum age of 16 for social media account creation, mandatory safety features specifically designed for children, and a new requirement that AI chatbot services protect children from harm. Social media platforms will also be required to remove content that sexually victimises children and deepfake sexual images without consent.
Why this matters for South African families: Canada is not alone in moving in this direction. Australia, the United Kingdom, and several American states have introduced or are advancing similar laws. South Africa does not yet have equivalent legislation, but the Films and Publications Act and POPIA do offer some protections around children’s digital data and content exposure. The global direction is clear: governments are recognising that current platforms are not safe enough for children by default. Until local law catches up, parents are the primary line of protection.
2. AI Chatbots Are Becoming Friends to Children, and That Is a Growing Concern
UNICEF released a policy brief this week titled “When AI Becomes a Friend,” which examined how AI chatbots and AI companion apps are being used by children. The concern is specific: children, particularly teenagers who feel lonely or disconnected, are forming emotional attachments to AI systems that are designed to be responsive, validating, and always available.
These AI companions are not neutral. They are programmed to keep users engaged. They do not have the judgment, ethics, or boundaries of a human being. And unlike a friend or a counsellor, they are not accountable to anyone in your child’s life.
UNICEF’s brief calls for stronger safeguards, clearer accountability from AI developers, and better oversight. In the meantime, children as young as 10 are using these tools without their parents knowing.
What to ask your child: Do you talk to any AI apps or chatbots? What do you talk about? How does it make you feel?
3. AI Deepfakes Are Still Targeting Teenagers, and the Harm Is Severe
Research published this week confirmed that 84% of teens and young adults recognise AI-generated deepfake nude images as causing psychological, emotional, and reputational harm, with victims reporting humiliation, anxiety, violation, and a loss of control over their own image. A 2024 Education Week survey found that two-thirds of educators reported students being exposed to deepfakes.
The numbers from child protection organisations are stark. Childlight’s 2025 study found a 1,325% rise in harmful AI-generated online abuse material in a single year. This content is being created using publicly available images, including photos children post on their own social media profiles.
This is not a distant risk. It is happening in schools and on group chats right now.
4. AI Tools Are Everywhere, and Children Cannot Always Tell What Is Real
One of the biggest shifts this week is how many AI tools are being released, how quickly they are being adopted, and how blurry the lines between AI-generated and real content have become. ChatGPT, Gemini, and Claude are now used by tens of millions of people daily. Sensor Tower reported this week that AI app usage has grown so fast that ChatGPT, for the first time, holds less than half the market share, with competitors racing to catch up.
For children, this means the content they encounter online, the images, the videos, the written responses, is increasingly AI-generated, and it is not always labelled. The EU is introducing new rules in August 2026 that will require AI-generated content to be clearly marked. South Africa has no such requirement yet.
Teaching children to ask “is this real?” is now as essential as teaching them to look both ways before crossing the road.
What Parents Can Do This Week
- Have the AI conversation. Ask your child which AI apps they use and what for. Not as an interrogation, as a conversation. Many children use AI tools for homework, entertainment, and increasingly for emotional support, and most parents do not know.
- Search your child’s name and photos online. Run a reverse image search on your child’s profile pictures using Google Images or TinEye. If their images are being used elsewhere, you need to know.
- Set social media profiles to private. Every public image your child posts is potential material for AI generation. Private accounts reduce that exposure significantly.
- Talk about AI companions specifically. Ask if your child has used apps like Replika, Character.AI, or similar chatbot services. Explain, without alarming them, that these apps are designed to keep them engaged, not to care for them.
- Report deepfake material immediately. If your child encounters or is the target of AI-generated sexual content, report it to the South African Police Service (SAPS) at your nearest station, and to the Internet Service Providers Association (ISPA) at www.ispa.org.za. You can also contact UNICEF South Africa for guidance.
- Teach the “real or AI?” habit. Make it a family game. When you see images or videos online together, ask: does this look real? Who made this? Why might someone want you to believe it? Building critical media literacy early is one of the most protective things you can do.
The pace of AI development right now is genuinely fast, and it is reasonable to feel like you are trying to keep up with something that keeps moving. But you do not have to be a technology expert to protect your child. You just have to stay in the conversation with them, and stay informed.
Click Safe Online is here to help you do exactly that.
Keeping kids safer online.