Building Trust with AI: Expanding Mental Health Access to Underserved Communities

Mental health issues affect millions of people worldwide, yet access to adequate mental health care remains a significant challenge, particularly for underserved communities. These populations often face barriers such as stigma, lack of resources, cultural differences, and geographic isolation. As a result, many individuals do not receive the support they need, leading to worsened mental health conditions and a diminished quality of life.

AI has the potential to bridge the gap in mental health care by providing scalable, accessible, and personalized support. AI-driven tools can offer a range of services, from mental health assessments and therapeutic interventions to ongoing monitoring and support. By leveraging AI, mental health services can be extended to reach individuals who have historically been underserved, fostering greater equity in care.
For this reason, TuConsejería, with the support of the Inter-American Development Bank, created an AI-driven tool to provide access to remote mental health services for underserved populations, including Indigenous women, victims of gender-based violence, and returning migrants in Guatemala, Colombia, Mexico, and the United States.

During this session, attendees will have the opportunity to interact with TuConsejería’s AI-driven tool and learn how AI is delivering trust-based mental health services to underserved communities in Central America and Colombia. TuConsejería has partnered with civil society organizations to reach populations that may never have had the chance to use AI. This AI-driven tool is now providing mental health support through WhatsApp, the most widely used mobile phone application in the region (Mobile Growth Association, 2021). Additionally, the tool is available in most world languages, and is in the process of learning to offer its services in K’iche’, Q’eqchi’, and other Indigenous Mayan languages in Guatemala.

This session will be highly interactive, using a World Café format to leverage the collective and diverse wisdom of SOCAP attendees. The goal is to gain a deeper understanding of the barriers to access to mental health services, and explore how AI can build trust among users by offering culturally sensitive and trustworthy tools for the health sector and beyond.

The entire session will last approximately 50 minutes:

Setting the Stage (5-7 minutes)

We will begin by “Setting the Stage,” which includes a brief story based on TuConsejería’s experience in Central America and Colombia. The purpose is to highlight the disparities in mental health access and demonstrate how AI can help bridge that gap. We will also provide a brief overview of the World Café structure, explaining that it will consist of three rounds, where participants will switch tables after each round.
Each table will be asked to assign a “table host,” who will summarize key takeaways on a flipchart or digital board after each round. This ensures that new groups can build upon the ideas generated by previous discussions.

Proposed Round Structure (30 minutes total)

Our proposed round structure and goals for each round are as follows:

Round 1: Understanding Barriers to Mental Health Access and Building Trust in Underserved Communities (10 minutes)

The goal of Round 1 is to surface the challenges that underserved communities face when accessing mental health services and explore how AI can help address these issues. To facilitate this, each table will be assigned a role. For example:
– One table may take on the role of a migrant returnee experiencing depression and seeking a therapist.
– Another table may assume the role of a rural school teacher witnessing the effects of gender-based violence on her students and struggling to find mental health support, particularly given the stigma in her community.

The number of roles will depend on the number of tables and attendees, with 6-7 participants ideally seated at each table.

Round 2: Designing AI for Equity and Inclusion (10 minutes)

The goal of Round 2 is to explore ways to ensure AI respects cultural nuances and promotes inclusivity.

In this round, participants will continue in their assigned roles (e.g., a rural teacher) but will now focus on exploring solutions using TuConsejería’s AI-driven tool. They will test the tool live and evaluate whether it — and AI tools in general — demonstrate cultural sensitivity, trustworthiness, and accessibility for diverse populations.

Round 3: Building Trust, Ethical Considerations, and Privacy (10 minutes)

The goal of Round 3 is to identify key principles that can prevent bias, ensure data privacy, and maintain transparency.
Building upon insights from previous rounds, participants will discuss the ethical considerations and safeguards needed to build trust in AI-driven mental health services, using their assigned roles as a lens for the discussion.

Harvesting Insights (15 minutes)

After the rounds, each table will be invited to highlight 1-2 major insights that emerged during the discussions. These insights will be captured using visual aids such as Sticky Notes or a digital collaboration tool like Mural to ensure they are documented for further reflection.

Closing Reflection (5 minutes)

To conclude, we will invite all participants to reflect on the question: “What’s one action we can take to ensure AI-powered mental health solutions are both trusted and accessible for underserved communities?”

One of the expected outcomes of this session is to promote collaboration and partnerships beyond SOCAP. We hope that by the end of the closing reflection, participants will leave with actionable ideas and potential partnerships to increase access to AI-driven mental health services for underserved populations around the world.