Models & Releases

ChatGPT therapy trend: 71% of users turn to AI for emotional support

Over 1 billion people struggle with mental health, and AI chatbots are stepping in.

Deep Dive

The article from Euronews highlights a significant shift in how people use generative AI: emotional support and companionship have become the most common applications, surpassing personal life management or health advice. Data from Harvard Business Review, a 2025 Línea Directa study, and the AXA Mind Health Report show that across Mexico and Spain, large percentages of individuals—especially young people—are turning to ChatGPT for mental health issues like anxiety, break-ups, and decision-making. The World Health Organization notes over 1 billion people have a mental health disorder, creating a massive unmet need that AI is beginning to fill.

However, experts like psychologists María Fernanda Cruz Urquiza and Cecilia Paredes Aldama caution that while ChatGPT offers 24/7 availability, validation, and a judgment-free space, it cannot replace the human bond essential for therapy. It may help reduce anxiety superficially by organizing thoughts or planning conversations, but it doesn't address root causes. There are also risks: a Guardian investigation found AI systems sometimes provide incorrect health information. The EU's AI Act classifies health-related AI as high-risk, requiring strict oversight. The article concludes that AI can be a helpful first resource, especially for those who can't afford therapy, but should not be mistaken for professional treatment.

Key Points
  • Harvard Business Review reports emotional support as the top use case for generative AI, ahead of personal life management.
  • In Mexico, 71% of people use AI tools for mental health support (AXA Mind Health Report); in Spain, 24.8% choose digital self-diagnosis first.
  • Psychologists warn ChatGPT offers constant validation but lacks depth; a Guardian investigation found AI can spread incorrect health information.

Why It Matters

AI is filling a critical mental health support gap, but lacks human nuance and risks misinforming vulnerable users.

📬 Get the top 10 AI stories daily