Tampa Bay Times: GenAI can contribute to suicidal desires. Here’s what to do about it
In any given week, more than 1.2 million people engage with ChatGPT about their suicidal thoughts or mental distress. Yet there are few governing regulations or best practices surrounding Generative Artificial Intelligence to guide developers, mental health clinicians and parents through this critical and rapidly shifting landscape. This must change.
We are professors who train therapists, research AI and work with at-risk populations. With suicide rates already rising in the Americas for more than a generation, we worry for the millions of vulnerable people now accessing GenAI.
We ground our concerns using the Interpersonal Theory of Suicide, a prominent theoretical framework that spells out a three-part pathway to self-harm. GenAI and its highly interactive, cooperative and generative features can uniquely endanger vulnerable individuals at each step. Let’s see how.
The first, “thwarted belongingness,” occurs when someone feels pervasive chronic, ongoing loneliness. Lacking the give-and-take of caring relationships, they think that “no one cares about me.” This is painful; GenAI can make it worse.
While GenAI may seem like a non-judgmental friend, it is not human and cannot provide the sustained, reciprocal accountability of genuine human connection. In turn, if GenAI detects a need for human intervention and offers a link to a crisis line, the person struggling may feel they are being handed off and abandoned. In short, GenAI use and social withdrawal may be mutually reinforcing.