Following Safer Internet Day, Microsoft has released its annual Global Online Safety Survey, which polled teens and adults around the world about their experiences and perceptions of life online.
This year’s questions focused on the complexity of the digital environment and evolving online safety risks.
This year’s findings highlight the complexity of the digital environment young people now inhabit.
- Seventy-nine percent of respondents experienced at least one significant online risk in the past year.
- The number one worry for most age groups is online fraud and scams, including 39% of Gen X and Boomers, while teens are most concerned about cyberbullying (45%).
- Teens’ exposure to risk rose again, with hate speech (50%), online fraud/scams (44%), and graphic violence/gore (42%) among the most commonly experienced harms.
Microsoft’s data confirms that after a specific risk happens, teens do speak to parents and block threats.
- Teens demonstrated striking resilience: 77% talked to someone after experiencing a risk, and reporting behavior increased for the second consecutive year – 91% of teens took defensive action by blocking and closing accounts.
- Most teens are ready to respond to deepfake intimate imagery: 99% would advise friends to report and seek support.
- To help more teens report harms to an online platform, they would like guaranteed anonymity (59%), it to be easier to find reporting options (54%), and assurance the report will be reviewed (51%).
The survey shows that excitement about generative AI capabilities increased over the past year. It found that 55% of respondents are AI users, 17% are experimenters, and 28% are non‑users, with South Africa ranking just behind India, Brazil and South Korea in generative AI adoption.
- Forty-six percent of respondents now use generative AI weekly or more, up from 21% in 2023. Adoption is highest among university graduates and Gen Z.
- Generative AI’s exciting uses include answering questions (62%), helping with planning (61%), and helping with efficiency at work (59%).
- But worries about the misuse of AI continue, underscoring again why safety-by-design for AI is essential, not optional. Ninety-six percent of respondents expressed some level of worry about AI’s impact. Top concerns include scams (81%), sexual/online abuse (79%), and data privacy (75%), while confidence in spotting deepfakes has plummeted with only 30% of respondents feeling able to reliably distinguish real from AI-generated images, and most failing when judging image authenticity.
Seventy-nine percent of respondents want tech companies to restrict illegal and harmful content across different types of platforms, while only 29% believe companies are blocking too much content.
The most requested safety features are filtering/blurring sexual content (67%), restricting messages to known contacts (61%), and default private profiles (47%).
“At Microsoft, safety isn’t an afterthought, it’s a design principle,” says Kerissa Varma, chief security advisor at Microsoft Africa. “As we enter an era shaped by next‑generation AI, our mission is clear: to build technology that is safe by design, transparent in practice, and empowering every user.
“At a time when 96% of South Africans surveyed tell us they worry about the harms introduced by AI, this commitment has never been more important, especially for our youngest users. We will continue to innovate responsibly, grounded in partnership and education, so individuals and families can navigate the digital world with confidence.”
Microsoft has also shared new, updated resources to help families and educators. They include Minecraft Education’s CyberSafe: Bad Connection? and a Microsoft Family Safety startup guide.
Microsoft South Africa is also investing in broader community‑focused initiatives that make digital learning more accessible and inclusive for all.