CRII: SaTC: Privacy vs. Accountability--Usable Deniability and Non-Repudiation for Encrypted Messaging Systems

全球数十亿人在使用端到端加密消息应用程序。然而，加密本身并不能保证人们的隐私、责任和安全。有时，安全目标需要合理的推诿：对于可能存在妥协的对话，发送一条消息并让接收者知道是谁发送的，但无法向其他人证明发送者是谁，这是有用的。有时问责目标需要不可否认性：对于记录对话，发送方可能希望能够证明他们或他们的合作伙伴发送了消息，即使合作伙伴声称不这样做并试图更改对话记录。这个项目的目标是帮助人们理解并做出正确的选择，什么时候选择可否认性，什么时候选择不可否认性，并开发消息应用程序的功能，使这些选择变得容易。通过这一点，该项目将提高人们如何理解和使用密码学的知识，以及人们有效和安全沟通的实际能力。该项目围绕两个并行的接口设计活动进行组织，一个关注可否认性的接口，另一个关注不可否认性的接口。对于可否认性，该团队将开发交互技术，允许用户更改文本中消息的内容、时间和作者，并利用支持这些编辑的底层加密技术。对于不可否认性，该团队将开发交互技术，帮助人们使用与其身份绑定的密钥对消息进行加密签名。这两个界面都将通过一系列用户研究进行测试，这些研究将解决(1)人们使用工具的能力，(2)他们理解使用工具的可否认性和不可否认性的含义的能力，以及(3)在会话环境中使用它们。这项工作将由一些本科生和研究生以及研究实习生进行，为未来的网络安全劳动力提供宝贵和必要的培训。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

End-to-end encrypted messaging apps are used by billions of people, globally. Encryption by itself, however, does not guarantee people’s privacy, accountability, and safety. Sometimes safety goals require plausible deniability: for conversations that might be compromising, it can useful to send a message and have the receiver know who sent it, but not be able to prove who the sender was to others. Sometimes accountability goals require non-repudiation: for on-the-record conversations, a sender may want to be able to prove that they, or their partner, sent a message even if the partner claims otherwise and tries to alter the conversation transcript. This project’s goal is to help people understand and make good choices about when to choose deniability and when to choose non-repudiation and develop messaging app features that make it easy to make those choices. Through this, the project will advance knowledge around how people make sense of and use cryptography, and people’s practical ability to effectively and safely communicate. The project is organized around two parallel interface design activities, one focused on interfaces for deniability, and one on interfaces for non-repudiation. For deniability, the team will develop interaction techniques that allow users to alter the content, timing, and authorship of messages in a transcript, leveraging underlying cryptographic techniques that support those edits. For non-repudiation, the team will develop interaction techniques that help people cryptographically sign messages using keys tied to their identity. Both interfaces will be tested through a series of user studies that address (1) people’s ability to use the tools, (2) their ability to understand the implications of using tools for deniability and non-repudiation, and (3) to use them in conversational contexts. The work will be carried out with a number of undergraduate and graduate students, and research interns, providing valuable and needed training for a future cybersecurity workforce.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.