Towards secure and trustworthy crowdsourcing: challenges, existing landscape, and future directions

Crowdsourcing has long been an effective way for data collection. Despite numerous research efforts devoted to crowdsourcing security, most crowdsourcing platforms in practice are still storing and processing plaintext data, which is vulnerable to in-house attacks. To fill this gap, we envision a full-fledged crowdsourcing platform with end-to-end encryption (E2EE), where the crowdsourced data remains always encrypted to the platform. Such a design would serve as an in-depth defense strategy against data breaches from both internal and external threats and provide technical means for crowdsourcing service providers to meet various stringent regulatory compliance. To realize this vision, we will discuss the technical requirements and related challenges, including: (1) securing worker profiles in task matching to respect individual privacy, (2) assuring high-quality crowdsourced data to enhance data values, (3) enabling versatile data analytics to uncover data insights, (4) protecting data at the front-end to fully achieve E2EE, and (5) preventing the abuse of E2EE for practical deployment. We will overview the limitations of prior arts in meeting all these requirements and identify potential research directions for the roadmap ahead.