The modern landscape of social care is shifting from a series of disconnected, manual processes to a more integrated, data-driven ecosystem. Leading this transformation is an expert in healthcare technology strategy who has dedicated their career to advancing interoperability and connected care. With deep experience in managing complex social care networks, our expert understands that while technology provides the tools, the heart of the work remains the human connection. They advocate for a model where artificial intelligence serves as the infrastructure rather than the architect, ensuring that no individual falls through the cracks of a fragmented system.
The conversation explores how technological integration can address the chronic shortage of capacity in social care without sacrificing the compassion that defines the field. We discuss the transition from manual intake and repetitive paperwork to automated systems that “connect the dots” between clinical records, social needs, and community resources. A significant portion of the dialogue is dedicated to the ethical considerations of using algorithms, specifically focusing on data messiness, the danger of AI acting as an automated gatekeeper, and the vital role of federal transparency rules. Ultimately, the discussion emphasizes that the future of care is not about replacing humans but about empowering them to move faster and see patterns that were previously hidden in the noise.
Social care is deeply human work, yet coordinators are often bogged down by manual intake forms and fragmented systems. How do you see AI bridging the gap between clinical data and the lived experience of a person facing housing or food insecurity?
Right now, we are dealing with a system where people are forced to tell their most painful stories over and over again because clinical, claims, and social data live in separate silos. AI has the unique ability to act as the connective tissue, identifying patterns across these fragmented data points to create a “whole person” picture that has been missing for years. Instead of a social worker spending hours manually cross-referencing referral lists or summarizing notes from three different agencies, these tools can synthesize that information in seconds, highlighting that a patient’s missed appointment isn’t a lack of compliance, but a transportation gap. It allows us to move beyond the paperwork and actually look the person in the eye, knowing their history of utility shutoffs or food insecurity is already integrated into their care plan. This isn’t about a chatbot offering advice; it’s about making the entire infrastructure responsive enough to catch someone before they end up in the emergency room.
We often hear that social care isn’t short on compassion, but it is desperately short on capacity. In what specific ways can technology act as a force multiplier for community health workers who are currently overwhelmed?
The real promise of this technology lies in its ability to handle the administrative heavy lifting that currently eats up about half of a care coordinator’s day. When we use AI to support closed-loop referrals and improve follow-ups, we are essentially giving that worker back the time they need to build trust with their clients. For instance, instead of a staff member calling twenty different organizations to find a housing opening, an intelligent system can surface real-time gaps in service availability and suggest the smartest match for that specific person’s needs. It transforms the role from a data entry clerk to a high-impact advocate who can focus on the nuance of behavioral health or caregiver strain. By reducing administrative waste and deciphering outcomes across complex programs, we are stretching our limited capacity much further than we ever could with manual spreadsheets and phone calls.
There is a significant concern regarding “false precision,” where a model might look certain but miss the nuances of a person’s life. How can organizations prevent AI from becoming a cold gatekeeper to essential services?
The danger of false precision is one of the most sobering risks we face because a model might show a resource is available when, in reality, the provider has no actual capacity to help. We have to be incredibly careful that an algorithm doesn’t become a barrier that decides who is “likely enough” to benefit from a service, effectively denying food or housing based on a historical data point that may no longer be true. For example, a person might screen negative for food insecurity today, but a sudden job loss or medical emergency next month changes everything; if the AI treats its initial assessment as permanent, that person is left in the dark. We must treat these recommendations as suggestions that require human vetting, ensuring that the real barriers—whether they are fear, language, or a lack of trust—are recognized by a person, not ignored by a machine.
Regulatory frameworks like the ONC’s HTI-1 and HHS Section 1557 are now central to the conversation. How are these mandates changing the way we implement predictive algorithms compared to the less regulated approaches of the past?
The shift toward high-level scrutiny is a necessary evolution because it moves AI from a “black box” to a transparent part of the certified health IT infrastructure. Under the ONC’s HTI-1 final rule, there are now strict transparency requirements that force developers to show how their predictive algorithms actually work, which is a massive leap forward for accountability. Similarly, the HHS Section 1557 rules are vital because they explicitly prohibit discrimination through patient decision support tools, ensuring that we aren’t just scaling up the biases already present in messy historical data. These regulations mean that in 2026, we cannot simply let a model run in the background without clear consent, privacy protections, and a way for humans to question or correct a decision. It forces us to build responsibility into the code from day one rather than trying to fix a biased system after the harm has already been done to vulnerable communities.
With the messiness of social care data—where those most in need are often the least represented—how do we ensure that innovation doesn’t inadvertently worsen the disparities it aims to solve?
Addressing bias in health AI is not a theoretical exercise; it is a daily commitment to auditing our systems and keeping “humans in the loop” at every stage. We have to acknowledge that if a dataset is incomplete or reflects decades of systemic inequity, the AI will naturally repeat those patterns unless we actively intervene. This means engaging community partners to validate what the data is telling us and being honest about the fact that technology cannot manufacture the trust required for social care to work. The organizations that succeed are those that treat AI as a tool for orchestration and connectivity, rather than a “rip and replace” solution for the systems people already use. We must ensure that the output of these models is always tied to real-world outcomes and that we never turn deeply personal social needs into just another layer of surveillance without a clear benefit to the individual.
What is your forecast for the integration of AI in social care over the next two years?
My forecast is that we will move away from the “shiny object” phase of AI and transition into a period where it becomes a quiet, essential part of our social care infrastructure. From 2026 to 2028, we will see a major shift toward automated coordination that finally bridges the gap between community-based organizations and health plans, making the ecosystem significantly easier to navigate for the average person. We will likely see a reduction in “referral loops” where people get lost in the system, thanks to more intelligent routing and better measurement of which interventions actually improve lives. However, this progress depends entirely on our willingness to build these systems responsibly; if we prioritize speed over transparency, we risk breaking the very trust that social care is built upon. The real winners will be the organizations that use technology to empower their staff to be more human, not less.
