Every year, innovators, researchers, and trust & safety professionals from around the world come together to participate in the T&S Hackathons — events dedicated to building safer digital ecosystems for everyone. The mission is simple but ambitious: to identify and build innovative solutions that protect users no matter where they roam online.
To achieve this, hackathons are held both remotely and across various global locations, ensuring that every voice — from developers to educators, from policy experts to designers — has the opportunity to share ideas that can truly make the internet a safer place.
If you’d like to contribute or follow ongoing discussions, you can join the T&S Hackathon on LinkedIn and connect with the global community of digital safety innovators.
The T&S Hackathon Idea List
We’re always generating fascinating and impactful ideas at T&S Hackathons! If you’d like to learn more about any of the projects below — or be connected to the teams behind them — reach out anytime at hello [at] tshackathon.org. Ideas appear in no particular order, and you can use filters on the official site to explore them by topic or event.
Highlighted Ideas from T&S Hackathons
AgeGuard
Develop an LLM to predict the age of a user with high confidence. A verified dataset would be built using parental control features (like Instagram Family Center or Google Family Link) to analyze behavioral factors such as login times or network connections. If the predicted and reported ages differ, certain content — such as political or sexually explicit material — would be restricted. Topic: Child Safety | Event: San Francisco 2023
Undersight Board
A “Voice of the Community System” for policy development and localization. This LLM-powered persona-building system helps moderators, policy managers, and operations teams generate community-specific input for Trust & Safety policy creation. Topic: Governance | Event: San Francisco 2023 | Top Idea
TrustShare
A generative chatbot designed for trust professionals, trained on publicly available trust & safety resources, transparency reports, and regulatory data. It serves as a knowledge companion for policy experts. Topic: Governance | San Francisco 2023
Safe Speak
An AI chatbot that helps users manage distressing situations online. It prompts users to report harmful content and connects directly to existing reporting workflows. Topic: User Education | San Francisco 2023
Safety Interventions
Combining behavioral analytics, fraud detection models, and textual analysis to detect scams or fraud signals in user interactions. The system provides real-time warnings and prevention education tailored to each platform’s needs. Topic: Scams & Fraud | San Francisco 2023
Clarif.ai
An AI-driven tool optimizing how platforms communicate safety messages. By learning about each user’s comprehension level and product usage, it delivers tailored, easy-to-understand guidance. Topic: Policy | San Francisco 2023
SafetyGPT
A web-based interface that provides users with simple, reliable safety information across major online platforms — including policy explanations and community guideline summaries Topic: Policy | San Francisco 2023
UnsafeSets
A synthetic dataset generator for training models on harmful content. It addresses the scarcity of real-world harmful content datasets, enabling safer, more adaptive AI moderation systems. Topic: Harmful Content | San Francisco 2023
AI Safety Network
A cross-platform abuse-sharing model powered by AI, allowing industries to create central clearing houses for hashed content related to abusive behavior. The initiative aims to unite public, private, and civic partners around privacy-first data standards. Topic: Data Sharing | San Francisco 2023
SafetySherpa
An educational chatbot for social media that teaches users about policies and potential violations when they engage with risky content. Topic: User Education | San Francisco 2023
XCALibR
A collaborative project to unify online content standards. The Cross-Platform Online Safety Collaboration Software uses machine learning to harmonize enforcement policies, emphasizing transparency and inclusivity. Topic: Data Sharing | San Francisco 2023 | Top Idea
TrustAlign
An AI tool simplifying the user agreement process by comparing personal values with platform policies. It highlights differences in simple terms, informing users before they violate rules — promoting accountability and transparency. Topic: Transparency | San Francisco 2023
One-Stop Policy Shop
A centralized resource leveraging public platform policies, academic research, and regulations to help build safer digital environments. The project uses multilingual LLM translation and “red teaming” to ensure accuracy. Topic: Policy | San Francisco 2023 | Top Idea
B.L.O.B.I. (Big List of Bad Ideas)
An AI model detecting emerging hate groups by learning from existing hate speech datasets. Updated in collaboration with NGOs and researchers, it offers real-time insights into harmful online movements. Topic: Data Sharing | San Francisco 2023
WatchDog
An info-sharing consortium allowing platforms to share behavioral signals of bad actors. The system develops AI models to detect and enforce actions against harmful networks, continuously evolving with new data. Topic: Data Sharing | San Francisco 2023
AI Intervention
AI-powered detection of scam activity in payment and e-commerce systems, proactively identifying suspicious behavioral traits. Topic: User Education | San Francisco 2023
AlgoSpeakEasy
A multilingual AI model detecting moderation evasion tactics across platforms. It categorizes behaviors, labels evasion types, and enables real-time updates through collaborative, privacy-safe APIs. Topic: Data Sharing | San Francisco 2023 | Top Idea
AI Gold Standards
A framework for evaluating AI systems to reduce bias, enhance contextual understanding, and improve fairness. Topic: Governance | San Francisco 2023
Chippy
A prosocial AI chatbot that helps prevent negative online experiences by promoting education, empathy, and engagement. It empowers users to take responsibility for community well-being while boosting trust and user satisfaction. Topic: User Education | San Francisco 2023
360.ai
A workplace AI platform analyzing text, speech, and sentiment to identify exclusionary behaviors. Helps HR departments ensure compliance with diversity, equity, and inclusion (DEI) standards. Topic: User Education | San Francisco 2023
SafeTrace Collaborative
An open-source AI repository uniting multiple social media platforms to detect and prevent child exploitation and trafficking. It maps patterns, shares insights, and strengthens global child protection efforts. Topic: Data Sharing | San Francisco 2023
Salama Button
A safety button designed to make abuse reporting faster and more accessible, especially for minors in distress. It simplifies the reporting process, acknowledging how fear impacts user behavior. Topic: User Education | San Francisco 2023
Acorn
An AI safety plug-in for young users that educates and protects children online through adaptive design. It grows with the user, offering age-appropriate guidance and safety recommendations. Topic: User Education | Paris 2023
Unified Safety Platform
A platform democratizing access to trust & safety intelligence, providing actionable regulatory information and plug-and-play tools for content monitoring. Topic: Data Sharing | Paris 2023
Think Twice
An AI chatbot that offers users real-time advice to prevent guideline violations. Before posting risky content, users receive context-aware alerts aligned with community policies Topic: User Education | Paris 2023
Policy Mapper
A benchmarking tool for policy teams at smaller platforms, bridging the gap between law and implementation by comparing content policies with legal standards. Topic: Policy | Paris 2023
Safe Sharing
A collaborative system reducing sexual harassment recidivism through cross-platform education and rehabilitation. Offers awareness programs for offenders and early detection of emerging threats. Topic: Data Sharing | Paris 2023
easyFlag
A centralized reporting portal connecting users with regional “trusted flaggers,” making it easier and faster to report illegal or abusive content. Topic: Governance | Paris 2023
Genuine
An independent governance model — a public-accountable board ensuring transparency and continuous innovation in digital risk management. Topic: Governance | Paris 2023
Nudge
An AI-powered real-time scam detection system focusing initially on sextortion cases. When a suspicious message is detected, users receive alerts, educational resources, and links to prevention tools. Topic: AI / ML | Paris 2023 | Top Idea
V Game
A gamified approach to content moderation where users flag violations to earn points and train algorithms — turning community engagement into safer digital spaces. Topic: Content Moderation | Paris 2023
AI Safety Education Initiative
An initiative to create and maintain modular educational materials about AI safety and digital well-being. Topic: User Education | Paris 2023
CrossGuard
An open-source tool integrating across social platforms to identify abusive accounts using AI-powered risk assessments, helping users avoid malicious actors. Topic: Data Sharing | Paris 2023
TrustNet
A knowledge-sharing platform for T&S professionals to exchange intelligence on new online threats, fostering early detection and collective response. Topic: Data Sharing | Paris 2023 | Top Idea
Conclusion: Innovation for Safer Digital Spaces
Each of these projects represents a commitment to making online spaces safer, more transparent, and more humane. Whether through AI-powered moderation, global data collaboration, or user education, T&S Hackathons show that when technology and ethics meet, digital trust grows stronger.
These ideas are not just prototypes — they are blueprints for the future of a safer internet.