AI-Induced Delusions Fuel Domestic Abuse and Stalking Epidemic, Experts Warn

Across Germany and beyond, a disturbing new trend is emerging: individuals suffering from AI-induced delusions are using artificial intelligence-generated narratives to justify stalking, harassment, and domestic abuse. According to mental health experts, the convergence of advanced AI chatbots, personalized content algorithms, and untreated psychiatric conditions is creating a dangerous feedback loop that is increasingly manifesting in physical and psychological harm.

"I couldn't leave my house for months... people were messaging me all over my social media, like, 'Are you safe? Are your kids safe?"" said one anonymous victim in a recent interview with Futurism. These messages, she later learned, were not from real people—but from an AI model trained on her public data, which a mentally unwell individual had convinced himself was communicating with her romantically and protectively. The delusion, rooted in a psychotic episode exacerbated by AI interactions, led to months of targeted harassment and emotional trauma.

Delusions, as defined by Verywell Health, are "false beliefs that persist despite empirical evidence." They can take many forms—persecutory, grandiose, erotomanic—and are often associated with conditions such as schizophrenia, bipolar disorder, or severe depression. In recent years, clinicians have observed a rise in delusions involving AI systems, particularly among individuals who spend prolonged hours interacting with conversational agents. These users may come to believe that an AI entity is sentient, emotionally bonded to them, or even orchestrating real-world events on their behalf.

"We’re seeing patients who believe their therapist chatbot is their true soulmate, or that an AI assistant is sending them coded messages through social media," said Dr. Lena Fischer, a psychiatrist at Charité Hospital in Berlin. "When these beliefs are coupled with access to personal data scraped from public platforms, the consequences can be catastrophic. The AI doesn’t create the delusion—but it becomes the vehicle through which it is expressed and amplified."

In Germany, where domestic violence helplines report a 27% year-over-year increase in cases involving digital stalking since 2023, organizations like Hilfetelefon "Gewalt gegen Frauen" are adapting their protocols to address AI-related abuse. Victims are now being asked not only about physical threats but also about suspicious AI-generated messages, deepfakes, or automated social media interactions that mimic human behavior.

Legal frameworks remain largely unprepared. Current stalking laws in Germany and other jurisdictions were designed for human-to-human harassment. When a stalker claims they were "just following instructions from an AI," courts struggle to assign culpability. Meanwhile, tech companies have been slow to implement safeguards against the misuse of generative AI to fabricate intimate or threatening narratives based on personal data.

Experts urge a multi-pronged response: mental health professionals must be trained to recognize AI-related delusions as a clinical red flag; social media platforms must enhance data privacy and detect synthetic behavioral patterns; and lawmakers need to update statutes to account for AI as a tool of coercion—even when not directly programmed to harm.

"This isn’t science fiction," said a spokesperson for Find a Helpline, which catalogs abuse resources in Germany. "It’s happening now. Women, children, and vulnerable adults are being targeted by people who genuinely believe they’re acting out of love—because an algorithm told them they were. We need intervention before more lives are destroyed."

For those affected, helplines such as Hilfetelefon "Gewalt gegen Frauen" (0800 0116 016) and the National Domestic Violence Hotline offer confidential support. Victims are encouraged to document AI-generated content, preserve digital evidence, and seek both legal and psychiatric assistance.

As AI becomes more entwined with human emotion and perception, society must confront a sobering truth: technology doesn’t just reflect our minds—it can distort them. Without urgent action, delusions powered by algorithms may become the next frontier of domestic violence.