Defending Against AI Threats with a User-Centric Trustworthiness Assessment Framework

This study critically examines the trustworthiness of widely used AI applications, focusing on their integration into daily life, often without users fully understanding the risks or how these threats might affect them. As AI apps become more accessible, users tend to trust them due to their convenience and usability, frequently overlooking critical issues such as security, privacy, and ethics. To address this gap, we introduce a user-centric framework that enables individuals to assess the trustworthiness of AI applications based on their own experiences and perceptions. The framework evaluates several dimensions—transparency, security, privacy, ethics, and compliance—while also aiming to raise awareness and bring the topic of AI trustworthiness into public dialogue. By analyzing AI threats, real-world incidents, and strategies for mitigating the risks posed by AI apps, this study contributes to the ongoing discussions on AI safety and trust.