Is your screen gaslighting you?
From viral conspiracy theories on TikTok to AI-generated voices mimicking loved ones, mental health professionals are raising the alarm on a disturbing trend: “AI psychosis”.
Researchers warn that AI chatbots and large language models can unintentionally reinforce delusions, sometimes triggering or worsening psychosis, especially in vulnerable users. Cases often follow a “slippery slope” from casual use to fixation, with themes of AI sentience, spiritual missions, or romantic attachment. These call for urgent safeguards and clinician-guided protocols to reduce these risks.
Julie Morrow, Chief Clinical Strategist at AddictionResource.net, works at the intersection of substance use, trauma, and mental health.
“We’ve crossed into a new kind of psychological vulnerability,” says Morrow. “It’s no longer just about screen time, it’s now about how advanced content generation can fracture someone’s sense of what’s real and what’s not.”
Tips on how to protect vulnerable users from AI-fueled paranoia through media literacy, screen-time limits, and early intervention:
- Set digital boundaries: Use screen time controls and limit exposure to algorithm-heavy platforms.
- Talk about media literacy: Especially with teens and vulnerable adults, explain how AI can mimic reality.
- Watch for changes: New paranoia, obsessive online behavior, or distrust in real-world interactions can be early signs.
- Don’t dismiss delusions: Validate concerns first, then seek professional guidance, especially if AI is being blamed.
- Involve mental health professionals: Early intervention matters, particularly for those with a history of psychosis or trauma.
“What concerns us most is how accessible these technologies are. You don’t need to be on the dark web to experience something destabilizing because you just need TikTok, ChatGPT, or a filter that manipulates your reflection. For those with schizophrenia, PTSD, or severe anxiety, it can tip them over the edge,” Morrow explains.
