AI may detect suicide risk early, but people aren’t ready to rely on it

New research shows that public acceptance of artificial intelligence (AI) systems for suicide prevention depends less on their technical capability and more on how much people trust them.

The study, titled "Investigating acceptance of current and future artificial intelligence systems for suicide prevention," published in AI & Society, sheds light on how people perceive both existing and future AI systems designed to detect and prevent suicidal behavior, and identifies the key factors that influence their acceptance.

The study notes that technological advancement alone is not enough. Public willingness to accept and trust these systems will ultimately determine their success.

Trust outweighs performance in shaping AI acceptance for mental health

The study is based on a large-scale survey involving participants from the Australian public, who were presented with a series of hypothetical scenarios involving AI systems used in suicide prevention. These scenarios covered both current forms of artificial intelligence, often referred to as narrow AI, and more advanced, future systems resembling general artificial intelligence.

Participants evaluated these systems across several dimensions commonly associated with technology adoption, including perceived usefulness, ease of use, social influence, available support, and trust. The findings reveal a clear pattern: while multiple factors influence acceptance of current AI systems, trust emerges as the single most important determinant when it comes to more advanced, future AI technologies.

For existing AI systems, acceptance is shaped by a combination of factors. People are more likely to support systems they believe will perform effectively, are socially endorsed, and can be trusted. However, when participants considered more advanced AI systems with greater autonomy and decision-making capabilities, trust became the dominant factor, overshadowing all other considerations.

This shift reflects a deeper psychological response to the perceived capabilities of AI. As systems become more powerful and less predictable, concerns about control, reliability, and ethical behavior intensify. In such cases, trust becomes the central lens through which people evaluate whether the technology should be used at all.

The study also finds that overall acceptance is significantly higher for current AI systems than for future, more advanced ones. This suggests that while the public is open to AI playing a role in suicide prevention, there is still considerable hesitation about granting greater autonomy to these systems.

Perceived benefits highlight potential for early intervention and support

The study further identifies a range of perceived benefits that contribute to public support for AI in suicide prevention. Participants recognize the potential for AI systems to enhance mental healthcare by enabling earlier detection of risk factors, improving access to support, and complementing existing human-led interventions.

AI systems can analyze large volumes of data, including behavioral patterns and digital interactions, to identify warning signs that may not be visible through traditional methods. This capability is particularly valuable in cases where individuals may not actively seek help or may struggle to communicate their distress.

The research also highlights the potential for AI to reduce barriers to accessing mental health support. By providing continuous monitoring and immediate responses, AI systems can offer assistance at any time, potentially reaching individuals who might otherwise fall through the cracks of conventional healthcare systems.

Another key benefit identified in the study is the potential to encourage help-seeking behavior. AI systems can serve as an initial point of contact, guiding individuals toward appropriate resources and support networks. This role is especially important in reducing stigma and making mental health support more accessible.

Participants also acknowledge that AI could help alleviate pressure on overburdened healthcare systems by supporting professionals with data-driven insights and decision-making tools. In this way, AI is seen not as a replacement for human care but as a complementary resource that can enhance overall system capacity.

Risks, ethical concerns, and barriers to adoption remain significant

The study also draws focus to several perceived risks that contribute to public hesitation. Concerns about data privacy, system reliability, and the potential for misuse are prominent, particularly given the sensitive nature of mental health information.

Participants express unease about how personal data would be collected, stored, and used by AI systems. The possibility of data breaches or unauthorized access raises significant ethical and legal questions, especially in the context of vulnerable populations.

Another major concern relates to the potential impact of AI on the quality of mental healthcare. There is a fear that over-reliance on automated systems could reduce human interaction, which is often a critical component of effective mental health support. Participants worry that AI may lack the empathy and contextual understanding required to handle complex emotional situations.

The study also highlights broader societal concerns, including distrust in AI systems and fears about their long-term implications. Some participants view advanced AI as posing risks not only to individual users but also to society as a whole, reflecting ongoing debates about the ethical boundaries of artificial intelligence.

These concerns are more pronounced when participants consider future AI systems with greater autonomy. This reinforces the finding that trust is central to acceptance, particularly as AI becomes more sophisticated and capable. The authors point out that addressing these concerns will require careful design, transparent communication, and robust regulatory frameworks. Ensuring that AI systems are safe, reliable, and aligned with human values will be essential for building public confidence.