Skip to main content
Talk

AI for All: Building an Inclusive Future in Health Tech

Dr. Omar Costilla Reyes joined the “AI for All” health-tech panel at the CHCI Tech Summit in Washington, DC, on November 19, 2025. The panel discussed explainable AI, access to care, and bias mitigation.

Group photograph from the AI for All session at the 2025 CHCI Tech Summit, Washington, DC, November 19, 2025. Credit: Congressional Hispanic Caucus Institute; image courtesy of Dr. Omar Costilla Reyes.
Speaker badge for Dr. Costilla Reyes at the 2025 CHCI Tech Summit, Washington, DC, November 19, 2025. Credit: Image courtesy of Dr. Omar Costilla Reyes.
CHCI’s announcement graphic for the 2025 Tech Summit, held November 19, 2025, at Capital Turnaround in Washington, DC. Credit: Congressional Hispanic Caucus Institute; image courtesy of Dr. Omar Costilla Reyes.
The AI for All session in progress at the 2025 CHCI Tech Summit in Washington, DC, on November 19, 2025. Credit: Congressional Hispanic Caucus Institute; image courtesy of Dr. Omar Costilla Reyes. View image source.

Panel on inclusive health technology

The Congressional Hispanic Caucus Institute (CHCI), a Washington, DC organization best known for its leadership programs for Latino students and young professionals, held its 2025 Tech Summit on November 19, 2025, at Capital Turnaround in Washington. CHCI reported attendees from government, industry, and community organizations, with themes that included closing the connectivity gap, preparing a tech-ready Latino workforce, and setting guardrails for AI without slowing innovation. In the program, the afternoon breakout session “AI for All: Building an Inclusive Future in Health Tech,” sponsored by AARP, started from the observation that AI is changing medicine, from diagnostics and drug discovery to virtual care and predictive models, and that these tools could reduce disparities or, without safeguards, deepen them for Latino communities that already face barriers to care. Dr. Omar Costilla Reyes joined Congresswoman Lori Trahan, Gidget Benitez Lewis, Iliana Estévez, Monica Lopez, and Alejandro Roark on the panel.

The ideas the panel discussed

In his account of the discussion, the panel examined how explainable predictive models are beginning to detect risk trajectories before clinical symptoms appear, opening opportunities for preventive care in historically underserved communities. An explainable model shows which information led to its prediction, so that clinicians and patients can question it, and a risk trajectory describes how a person’s estimated risk changes over time. The panel also explored the role of telehealth and conversational AI in expanding access to mental-health and primary care where provider shortages limit continuity. For bias mitigation it stressed three practices: data provenance, meaning knowing where training data came from and whom it represents; transparent model evaluation; and continuous performance monitoring across demographic groups, which checks whether a tool works equally well for different populations. Congresswoman Trahan’s engagement underscored the need for policy frameworks that prioritize safety, accountability, and equitable deployment.

Connection to Equ Healthcare and later conversations

Dr. Costilla Reyes connected the conversation to his work at Equ Healthcare, which he described that morning as shaping policy, improving access, and advancing equitable innovation in healthcare AI. He wrote afterward that a future of healthcare built on trust, auditability, and equitable access depends on multidisciplinary partnerships between scientists, technologists, and legislators, and that the summit provided a valuable forum for that collaboration. Looking back in January 2026, he thanked the CHCI team for organizing the event and described the pleasure of meeting policymakers and the Washington, DC community. The same themes of explainability and trust continued in his December 2025 appearance on ZETA’s telehealth panel, and his earlier World Salon panel “From Lab to Cloud” covered related questions of privacy and algorithmic bias.

Topics