Skip to main content

The Trevor Project joins the Youth AI Safety Institute as a standards steering partner

The leading suicide prevention organization for LGBTQ+ young people will bring its direct service experience to the Institute's first AI safety standards.
Trevor project logo

By Kasey Flemer

Standards Program Director, Youth AI Safety Institute

 

The Trevor Project is joining Common Sense Media's Youth AI Safety Institute as a standards steering partner, alongside Crisis Text Line and the National Alliance on Mental Illness. The Trevor Project will help shape our standards for how AI products should respond to young people dealing with suicidal ideation and other severe mental health challenges, drawing on its extensive research and nearly three decades of direct-service crisis work with LGBTQ+ young people.

 

When young people are struggling, more of them are turning to AI—especially when they can't access identity-affirming human crisis care. For LGBTQ+ young people who may not feel safe talking to the people around them, that conversation can matter even more.

 

That's why we're glad to have The Trevor Project help steer our AI standards. When a young person discloses their identity and expresses a crisis, does an AI product connect them to human specialized support they can actually reach, or fall back on a one-size-fits-all response."

 

Counselors at The Trevor Project support LGBTQ+ young people in crisis around the clock, and they know what helps," said Robbie Torney, the Institute's Head of AI & Digital Assessments. "That real-world practice is what these standards need, because advice that's safe for one teen—like 'talk to your family—can put another at risk."

 

The Trevor Project provides free, 24/7, confidential crisis services to LGBTQ+ young people by phone, text, and web chat. It helped pilot the 988 Suicide & Crisis Lifeline's specialized services for LGBTQ+ youth in 2022.

 

"AI use is particularly pronounced among LGBTQ+ youth. According to our research, 53% of LGBTQ+ youth surveyed engage with AI platforms at least occasionally, nearly half have used AI for friendship or someone to talk to, and 40% have used AI for mental health support," said Tommy Marzella, Vice President of Social Platform Development & Safety at The Trevor Project. "A high-risk group for suicide, LGBTQ+ youth require specialized, tailored care—care that we know is best delivered by highly trained humans. However, it's absolutely essential that AI models are trained and programmed with those care standards so that LGBTQ+ youth are safe and protected while using AI-generated support."

 

The Youth AI Safety Institute is developing our first standards with guidance from steering partners and input from the field. We will publish them, and the methodology behind them, in the coming months.