A new research-informed approach to mental health support is seeking to address long-standing gaps in access to emotional care, particularly for communities often left out of traditional systems.
Karen Paul, a clinician and doctoral researcher, developed the conceptual framework behind LilyHealth after drawing on both personal experience and professional training in community mental health settings.
Paul’s early exposure to mental health struggles in her South Asian immigrant community, where psychological distress is often stigmatized and left untreated, shaped her view that conventional mental health services frequently fail to reach those most in need.
While working in community-based mental health contexts in San Francisco, Paul began asking a fundamental question: how might emotionally supportive resources be made more accessible to individuals who face long wait times, cost barriers, or cultural disconnects?
At the same time, she observed how artificial intelligence was being used in fields such as education and finance to increase access and personalization.
These observations informed the development of LilyHealth, a research-informed, AI-enabled mental health support framework named in honor of Paul’s late mother.
LilyHealth is designed as a non-clinical emotional support model, rather than a diagnostic or therapeutic system. Its purpose is to illustrate how culturally responsive principles drawn from evidence-based therapeutic frameworks could be translated into accessible, technology-enabled formats.
The framework is not intended to replace licensed therapists or formal mental health care. The conceptual model emphasizes guided reflection, coping strategies, psychoeducational content, and emotionally validating prompts, with particular attention to cultural context and trauma-informed language.
It is intended to complement, rather than substitute for, traditional mental health services. Paul has emphasized the importance of ethical safeguards, including bias mitigation, privacy protections, and professional oversight, as core design considerations within the framework.
Any visual materials or illustrative examples associated with LilyHealth are non-functional and are used solely to communicate the underlying concepts.
She sees this work as especially relevant for young people and individuals navigating stress, identity challenges, and systemic barriers to mental health care. In the future, Paul hopes the framework may inform collaborations with educational or community-based organizations focused on mental health access and equity.
Although artificial intelligence is not a cure-all, Paul argues that thoughtfully designed, culturally aware tools can help bridge gaps in emotional support for people who might otherwise never engage with mental health resources.
She views this work as part of a broader effort to advance more equitable and inclusive approaches to mental health care in the United States.
