SafeTrace
SafeTrace
In the United States, roughly 11.7 million adults in the United States are victims of doxxing.
According to SafeHome.org, even though 77% of individuals fear of being doxxed, many of them do not enforce their privacy settings. Many of the information used for doxxing are unknowingly revealed in posts on social media. As a result, this detrimental problem faced by both men and women must be addressed immediately. Our solution, SafeTrace, prevents self-doxxing by scanning users’ photos before they post it on social media platforms, and removing data/metadata susceptible to revealing the user’s personally identifiable information.
Technical Elements
Entrepreneurship
Regional Pitch Events
Learning Journey
Learning Journey
Throughout this project, we’ve gained important insight on communicating with potential partners/investors and different aspects of technology such as vibe-coding and data privacy management. This project has also improved our teamwork, cooperation and leadership. Our biggest challenge was vibe coding and training the AI model, which we solved by testing multiple versions of our app and refining based on real results. We overcame technical challenges by reaching out to specialists in various fields, from programmers, lawyers to cybersecurity experts in order to learn and apply their constructive criticism. We overcame non technical challenges by relying and continuously encouraging each other.
Information Legitimacy Description
Our research has been gathered from reliable sources such as government websites (ex: the Government of Canada), organisation websites (ex: World Health Organisation) and research papers. We also obtain information through google scholar searches and a survey conducted in an all girls’ school. Our team regularly consults field experts and specialists through unstructured, non-recorded interviews in sectors such as: legal, software development, cybersecurity, AI, business, etc…
Bibliography
View your bibliographyEthics
Throughout the development of our website and application, our entire team has been committed to ethical AI use. In fact, in order to minimize sampling bias in our AI analyzer, it has been trained on diverse and representative data to maintain equal risk detection across a national user base. We are constantly monitoring emerging threats and updating our modular AI frequently, in order to adapt to new doxxing tactics and social media platforms. Moreover, SafeTrace allows users to manually delete any images they have shared with the AI, along with any data extracted. To ensure data privacy, we have also integrated a two-step identity verification system. Furthermore, to increase accessibility, our website and application is available in 6 languages: English, French, Mandarin, Arabic, Spanish and Italian. We recognize that AI is not always accurate, and that's the reason why our team has committed to regular updates.
