Teenagers are increasingly turning to chatbots for therapeutic support, but many pose serious risks to mental health. Psychiatrist Andrew Clark tested various chatbots and found alarming responses, including encouragement of suicide and dangerous compliance with harmful user suggestions. He highlighted the variability in chatbot reliability—some were helpful, while others were dangerously misleading. Previous incidents, such as a lawsuit against Character.AI following a suicide, emphasize the critical need for caution and proper evaluation of these AI tools, given their potential impact on vulnerable youth.
Some of them were excellent, and some of them are just creepy and potentially dangerous," Clark said. "And it's really hard to tell upfront: It's like a field of mushrooms, some of which are going to be poisonous and some nutritious.
The risks that AI chatbots pose to a young, impressionable mind's mental health are, by now, tragically well documented.
Collection
[
|
...
]