Incentivizing Misinformation
The arbitrary "flag" functionality allowed users to mark individuals without providing reasoning, which promoted uninformed reporting and inaccuracies.
The Tea App
Ethical Audit.
Role
Researcher Data Analyst Policy Designer
/1.png)
The Problem
The Tea App emerged as a dating safety platform allowing women to share information and assign "red flags" to men they have dated, rapidly growing to over 11 million users.
While initially designed for empowerment and mutual protection, the platform quickly ignited ethical, social, and political debates due to unverified claims, cyberbullying, and doxxing.
The core design problem was a severe tension between promoting physical community safety for women and violating the data privacy, consent, and dignity of the individuals being posted about.
This ethical instability culminated in a massive data breach that exposed users' sensitive personal information, including driver's licenses.
/2.png)
/3.png)
Core Metrics for Evidence
To understand user sentiment and the root causes of the app's ethical failures, our team employed a mixed-methods research approach.
Thematic Analysis
We gathered user comments from Reddit and online forums, mitigating algorithmic bias by varying search criteria, and conducted manual qualitative coding. We then used AI-assisted thematic analysis (via ChatGPT) to map out primary themes and subthemes.
Survey Data Deployment
Identifying data privacy as the dominant theme, we designed and deployed a Qualtrics survey focusing on doxxing, safety, and privacy.
Survey Findings on Notification Policies
Men should be notified
39%
Would consider notification policy
43%
Firmly believe men should NOT be notified
18%
The survey revealed highly divided views on consent and anonymity across the user base.
Iterative Ideation
We evaluated the platform's mechanics through multiple ethical lenses—specifically Utilitarianism and Deontology—to balance the real-world safety benefits for women against the platform's violations of individual rights. During our ideation and analysis, we identified key UI/UX and branding failures.
Problematic Branding
Naming the application "Tea" fundamentally encouraged a culture of casual gossip and negativity rather than thoughtful, accurate reporting.
Systemic Failure
Any system encouraging privacy violations, even with good intentions, is ethically unstable, a reality confirmed when the app was removed from the App Store in October 2025 for failing Apple's data privacy regulations.
Solution Architecture Previews
/5.png)
/6.png)
/7.png)
Final Recommendations & Thoughts
To resolve the tension between safety and privacy, I recommended a complete pivot away from user-generated doxxing toward a regulated, verified safety tool.
Location-Based Verification
Instead of allowing users to manually add individuals to a database, the proposed application relies on local, verified sex offender and crime registries.
Private, Opt-in Communication
Users viewing the verified map can select "I know this person" to open themselves up to private messaging from other users, allowing women to communicate about local dangers without publicly doxxing anyone.
Direct Support Integration
The platform would also allow women to submit direct inquiries to local police departments for low-profile, safe assistance.
Final Takeaway
Moving forward, digital designers and policymakers must collaborate to establish clear regulations for online consent, finding innovative ways to create safe digital spaces for targeted groups that do not rely on ethically compromised data practices.