Protects the LGBTQ+ Community with Proactive Moderation Efforts
TrevorSpace needed to move beyond manual moderation and user flags to protect its community at scale. Alice deployed contextual AI detection tuned to the LGBTQ+ community's language, with automated, policy-aligned enforcement across thousands of forums.
Lifesaving Support for LGBTQ Youth
Company Size
Industry
About
“As the leader of our peer-to-peer networks, it’s crucial to ensure that we are creating a safe online space for all LGBTQ+ young people. With Alice, we’ve found a partner from day one to help safeguard our community, while strengthening our real-time moderation efforts. ”
The Trevor Project is the leading suicide prevention and crisis intervention organization for LGBTQ+ young people. With the rise in online harassment against the LQBTQ+ community, The Trevor Project is a critical place for young LGBTQ+ folks to have as a safe resource to reach out to and feel protected. With its goal to prevent LGBTQ+ youth suicide and self-harm, it was vital for The Trevor Project to implement Trust & Safety mechanisms to proactively moderate messages that could be toxic or harmful in order to increase the protection of their community. By moving to a proactive approach, they were able to quickly take action on harmful content and spend more time building a thriving community focused on peer-to-peer engagement.
Challenge
When a young person reaches out on TrevorSpace, they may be in crisis. In the case of suicide and self-harm, it's crucial to ensure the team takes action quickly to help a user access life-saving care in a timely manner. That's the environment The Trevor Project operates in every day.
Given the volume of content posted to TrevorSpace on a daily basis, it's impossible to monitor every interaction on the platform. Prior to Alice, the team relied on manual moderation, historical systems, and user flags to catch content that violated their policies. As a non-profit with limited resources, this approach couldn't scale with the growth of the community.
When launching TrevorSpace, the team understood the importance of implementing safety by design. Yet after launch, the popularity of the site meant they needed a vendor to help reduce reliance on user flags and automate content moderation for specific abuse areas, taking action on harmful content in a more operationally efficient way.
How Alice Helped
Detection
The Trevor Project needed a content moderation vendor that could cover the violations critical to their community: harassment and bullying, hate speech, child solicitation, suicide and self-harm. Coverage alone wasn't enough. The quality of the models was crucial. In an effort to reduce undetected content, they turned to ActiveScore, Alice's contextual AI automated detection capabilities.
A key challenge was ensuring detection models understood the difference between harmful content and the language LGBTQ+ youth naturally use in their community. Alice customized its hate speech models to identify relevant keyword lists that would remove words commonly used among the LGBTQ+ youth community from being flagged, aligning detection to The Trevor Project's specific policy rather than applying a generic model.
Enforcement
To strike a balance between providing a safe space and allowing the necessary freedom for community expansion, The Trevor Project needed a partner to implement their warnings and penalties guidelines quickly and effectively on TrevorSpace.
When a user violates a guideline and the moderation team becomes aware of it, users are issued warning points. TrevorSpace leverages no-code workflows to automatically implement these policies. Anyone with 0-5 points automatically receives a warning. Anyone with 6-7 points receives a two-week suspension. Those with over 8 points are permanently banned. They also use moderation queue management to manually moderate community messages with greater efficiency.
The Results
By using Alice, The Trevor Project is able to ensure greater protection against the most egregious harms facing their community on the TrevorSpace platform. Detection models are tuned to their specific community rather than working against it. Enforcement is automated, consistent, and aligned to their policies without requiring manual intervention at every step.
By incorporating a proactive approach to moderation, they have moderated thousands of forums on the platform and ensured that their users have a safe space to discuss the issues that matter most to them.
Trusted by security and product teams in the world's most regulated industries
Alice brings years of adversarial intelligence expertise to AI security. We give enterprise teams the coverage that generic guardrails and one-time audits can't match.
Get a demo
What’s New from Alice
HIPAA Audit Is Just the Start
Passing a HIPAA audit doesn't mean your AI will behave safely in production. As healthcare AI takes on more complex roles in patient care and documentation, static compliance frameworks can't keep up with the behavioral risks that emerge in real-world systems. Here's how WonderSuite closes the gap.
Afraid AI Will Replace You? Here's the One Skill It Can't
James Villarrubia went from building AI for NASA's drone and aerospace programs to becoming CTO of a travel tech company. In this episode, he and Mo get into why curiosity might be the most important skill in the AI era, what happens to our brains when we stop pushing back on the answers we get, and why the people most resistant to AI might actually be seeing something the rest of us are missing.
It Takes AI to Break AI: The Case for AI Red Teaming
As AI systems gain autonomy, organizations need security approaches built specifically for AI behavior. Learn why AI-driven red teaming is becoming a critical defense layer.
Evaluation of Instagram Teen Accounts
This report evaluates default and opt-in content protections under real-world and adversarial conditions. The study examines safeguard effectiveness, resilience against attempts to surface inappropriate content, and platform improvements made following testing.