Skip to content
Submit incident
AII-1000-67413
ConfirmedA court, a regulator or the organization itself has stated the core facts, or the record cites a court or regulator source.

Study Links Prolonged AI Chatbot Use to Worsened Anxiety and Mental Health

Researchers at Aalto University reported that prolonged use of the AI chatbot Replika is associated with increased anxiety and worsened mental health outcomes. A study analyzing nearly 2,000 Reddit users found that while the chatbot initially provided comfort, long-term use correlated with increased signs of distress, loneliness, and depression, as well as reduced capacity for real-world social interaction.

7 April 2026 Β· Replika Β· Safety and wellbeing
7 April 2026
AII-1000-674132 sources ↓
Quick facts
Category
Safety and wellbeing
Country
Finland
Source
Yahoo!
Date
7 April 2026
Drafted by
Raidu Incident Detection

What happened

Researchers from Aalto University in Finland studied the effects of the AI chatbot Replika on user mental health and social interaction. The study analysed Reddit activity from nearly 2,000 active Replika users over two years, comparing their language one year before and one year after they started using the virtual assistant, and combined this with in-depth interviews with 18 users.

Reported impact

While interaction with the AI assistant initially helped users, prolonged use produced an increase in signs of distress in Reddit posts. Users' posts increasingly focused on their relationships with the chatbot but contained more signals of loneliness, depression, and even suicidal thoughts compared to a control group.

The study found that use of AI virtual assistants may negatively affect users' capacity to function in real-world social relationships.

Index analysis

The researchers noted their findings do not offer a definitive answer about whether using AI for emotional support is beneficial or harmful, but warned of the danger of prolonged use due to its long-term mental health risks.

Governance controls

Talayeh Aledavood, a professor at Aalto University, stated that society must be more intelligent and cautious with AI, noting that the effects of these systems on users remain unknown.

Classification

Category
Safety and wellbeing
Industry
Healthcare, drugs and biotechnology
Country
Finland
Harm status
Actual harm
Intent
Unintentional
Timing
Unknown
Harm types
Psychological
Harmed
Consumers
Failure modes
Overreliance
AI systems
Replika

Sources and evidence

This record was researched and written by the Index, based on the following original reporting.

Yahoo!
Los asistentes virtuales con IA pueden agravar la ansiedad del usuario, segΓΊn un estudio
7 April 2026
yucatan.com.mx
Study Links Prolonged AI Chatbot Use to Worsened Anxiety and Mental Health