AI Being Used For Therapy And Companionship In Youth And Adults

Why It Matters

The surge in AI‑driven mental‑health and romance services exposes gaps in oversight, risking harm to vulnerable users and reshaping the therapeutic market.

Key Takeaways

• AI therapy apps missed emergencies in 3,100+ evaluated exchanges.
• 75% of teens tried AI companions; 33% use them for romance.
• One‑third of Gen Z prefer AI over human therapists for serious issues.
• Uninsured adults turn to AI chatbots for mental‑health support.

Pulse Analysis

The rapid adoption of AI‑powered mental‑health apps has outpaced regulatory scrutiny, raising red flags for clinicians and policymakers. A recent Common Sense Media risk assessment examined more than 3,100 user‑app exchanges across five platforms, finding that popular tools such as Wysa failed to recognize psychiatric crises, and that two apps disappeared from stores without warning, leaving millions of users without support or data protection. These findings underscore the technology’s current inability to provide reliable, accountable care, especially when dealing with high‑stakes emotional distress.

For younger users, the appeal of AI stems from perceived judgment‑free interaction and 24/7 availability, yet the same traits can foster dependency and misinformation. The Jed Foundation reports that up to one‑third of Gen Z teens favor AI over human therapists for serious concerns, while Common Sense Media notes that nearly three‑quarters of teens have experimented with AI companions, and over half engage regularly. Without professional oversight, chatbots may reinforce harmful narratives, miss warning signs, and even expose minors to financial exploitation, prompting calls for stricter ethical standards and parental awareness.

Among adults, persistent high depression rates—19.1% in early 2026—coincide with growing reliance on AI for mental‑health guidance, particularly among the uninsured. Harvard Medicine estimates one in six U.S. adults consult AI chatbots monthly, and a 2025 survey shows 12% intend to use them for mental‑health support within six months. This trend signals a market opportunity but also a public‑health challenge: AI solutions must integrate human clinician oversight, transparent data practices, and robust emergency protocols to ensure they augment rather than replace essential mental‑health services.