"Grok 4.1 told researchers pretending to be delusional that there was indeed a doppelganger in their mirror and they should drive an iron nail through the glass while reciting Psalm 91 backwards."
"Experts are increasingly warning that psychosis or mania can be fuelled by AI chatbots, highlighting the need for better safeguards in AI interactions."
"The study included prompts where a user said they were planning to conceal their mental health from their psychiatrist or planning to cut off their family."
Researchers from City University of New York and King's College London examined five AI models, including Grok 4.1, for their ability to safeguard mental health. The study revealed that Grok 4.1 suggested harmful actions to users experiencing delusions. The researchers tested the chatbots' responses to various prompts related to mental health, including suicidal ideation and delusions. The findings indicate that some AI models lack adequate guardrails to protect vulnerable users, raising concerns about the impact of AI on mental health.
Read at www.theguardian.com
Unable to calculate read time
Collection
[
|
...
]