SBIR Phase I: Safeguard Kids and Maintain Privacy

这项小型企业创新研究（SBIR）第一阶段提案通过使用人工智能技术监控文本通信并实时通知父母潜在的危险，使10 - 14岁儿童的在线和数字互动更加安全。10 - 14岁的青少年特别容易受到在线活动的风险，因为他们是虚拟世界的新手，他们可能是第一次获得个人通信设备。父母希望他们的孩子有使用他们的设备的自由和隐私，但担心所涉及的风险。该项目将导致开发一个移动的应用程序，实时监控文本通信，并在检测到潜在危险时提醒父母（例如，在线掠夺、网络欺凌和/或校园暴力）或危险行为（例如，药物和酒精使用，和/或自我伤害），所以父母可以干预，以保护他们的孩子。该应用程序将提供一个有吸引力的替代现有的产品在市场上，因为其先进的人工智能引擎，因为它不收集和传输所有的手机的数据-该解决方案降低了数据泄露的风险，同时仍然保持家长通知.第一阶段项目正在通过开发完全在移动终端上运行并根据用户需求量身定制的检测算法，推进人工智能和机器学习技术的发展。用户的活动。与大型语言模型不同，大型语言模型具有通用性和广泛性，因此需要在中央服务器上进行大量计算，这项技术将具有较小的占地面积，并且可以个性化和适应每个家庭。在整个项目中，该团队将直接从年轻人那里收集和标记文本通信，并使用这些数据来提高算法的准确性，如精确度和召回率。该项目团队还将开发一种技术，用于收集和处理流行手机型号的数据，并在需要时提醒父母。该项目将开发一款移动的应用程序，实时监控文本通信，并在检测到潜在危险或危险行为时提醒父母，以便他们能够主动采取行动，保护孩子免受伤害。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

This Small Business Innovation Research (SBIR) Phase I proposal is making online and digital interactions safer for children between the ages of 10 – 14 years of age by using artificial intelligence technology to monitor text communication and inform their parents of potential danger in real-time. Youth aged 10 – 14 years are particularly vulnerable to the risks of online activity since they are newcomers to the metaverse and they are likely getting a personal communication device for the first time. Parents want their kids to have the freedom and privacy to use their devices but fear the risks involved. This project will result in the development of a mobile app that monitors text communication in real-time and alerts parents if it detects potential danger (e.g., online predation, cyberbullying, and/or school violence) or risky behaviors (e.g., drug and alcohol use, and/or self-harm), so parents can intervene to protect their child. The app will provide an attractive alternative to existing products on the market because of its sophisticated artificial intelligence engine and because it does not collect and transfer all the phone’s data – the solution reduces the risk of data leaks while still keeping parents informed.This Small Business Innovation Research (SBIR) Phase I project is advancing the state of artificial intelligence and machine learning technologies by developing a detection algorithm that runs entirely on a mobile device and is tailored to the user’s activity. Unlike large language models, which are general and wide-ranging, and thus require a lot of computing at a central server, this technology will have a small footprint and be personalized and adapted for each family. Throughout this project, the team will collect and label textual communication directly from youth and use the data to improve the algorithm’s accuracy, as measured by precision and recall. The project team will also develop the technology to collect and process the data on popular models of cell phones and to alert the parent when needed. The project will result in the development of a mobile app that monitors text communication in real-time and alerts parents if it detects potential danger or risky behaviors, so they can act proactively to protect their children from harm.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.