{"id":2331,"date":"2026-10-10T03:18:59","date_gmt":"2026-10-10T03:18:59","guid":{"rendered":"https:\/\/www.comforcare.com\/north-san-diego\/?p=2331"},"modified":"2026-10-07T18:43:59","modified_gmt":"2026-10-07T18:43:59","slug":"human-centered-ai-direct-care","status":"publish","type":"post","link":"https:\/\/www.comforcare.com\/north-san-diego\/resources\/blog\/human-centered-ai-direct-care\/","title":{"rendered":"Human-Centered AI in Direct Care: What Families and Care Teams Should Know"},"content":{"rendered":"\n

AI Is Entering Care Workflows<\/h2>\n\n\n\n

Artificial intelligence is becoming part of more workplace processes, including some areas of direct care and services that support older adults and people with disabilities.<\/p>\n\n\n\n

On September 29, 2026, the Administration for Community Living updated its resource hub focused on artificial intelligence and the direct care workforce. The collection brings together reports and educational resources examining how AI may affect services that help people live in their homes and communities.<\/p>\n\n\n\n

Potential uses include reducing administrative work, supporting workforce training, improving scheduling and coordination and helping teams organize information. At the same time, the resources highlight concerns involving privacy, bias, job quality, accessibility and the additional burden that a new technology can place on workers.<\/p>\n\n\n\n

Because human-centered AI in direct care<\/strong> is still an emerging area, families and care organizations should look beyond marketing claims. A technology company’s promise that a tool will save time, improve care, or work fairly does not by itself demonstrate that those outcomes will occur.<\/p>\n\n\n\n

\"human<\/figure>\n\n\n\n

Question 1: What Problem Are We Trying to Solve?<\/h2>\n\n\n\n

Technology decisions should begin with a clearly defined problem.<\/p>\n\n\n\n

A care team might struggle with duplicate documentation, missed handoffs, inaccessible training materials, scheduling challenges, or difficulty coordinating information. Identifying the specific issue makes it easier to determine whether a technology actually addresses the need.<\/p>\n\n\n\n

Without that step, an organization could introduce another system that creates additional data entry while leaving the original problem unresolved.<\/p>\n\n\n\n

Families can also ask practical questions. What will change in the person’s actual care experience? How will the organization know whether the change is helpful?<\/strong><\/p>\n\n\n\n

Question 2: What Information Does the Tool Use?<\/h2>\n\n\n\n

Care information can be deeply personal. Before an AI system is introduced, organizations should understand what information it collects, where that information is stored, who can access it and whether the data may be used to train another system.<\/p>\n\n\n\n

People receiving services deserve an explanation they can understand. Meaningful consent should not be hidden inside lengthy terms or complicated forms.<\/p>\n\n\n\n

Organizations also need clear procedures for responding to privacy concerns or potential data breaches.<\/p>\n\n\n\n

For families, caution is especially important when using general-purpose AI tools. Health information, financial information, identifying details and service-related information should not be entered into a general AI system unless an authorized organization has approved the use and clearly explained the applicable protections.<\/p>\n\n\n\n

\"human<\/figure>\n\n\n\n

Question 3: Who Can Correct an Error?<\/h2>\n\n\n\n

AI systems can produce inaccurate summaries, recommendations, classifications, or other outputs. Within a care setting, an incorrect result could potentially affect communication, scheduling, documentation, or service decisions.<\/p>\n\n\n\n

A human review process provides an important safeguard.<\/p>\n\n\n\n

People receiving support, authorized representatives, and direct care workers should know how to report an error and who has the authority to correct it. Important care decisions should not depend solely on an unexplained automated recommendation.<\/p>\n\n\n\n

This matters particularly for people who communicate in nontraditional ways, use augmentative or alternative communication, speak a language that a system handles poorly, or have records containing outdated information.<\/p>\n\n\n\n

Human review helps ensure that technology does not replace the context that caregivers and families know firsthand.<\/p>\n\n\n\n

\"human<\/figure>\n\n\n\n

Question 4: Does the Tool Work for the People Involved?<\/h2>\n\n\n\n

Accessibility should be evaluated with real users, not only through a technical demonstration.<\/p>\n\n\n\n

A digital system might meet an accessibility standard and still be confusing or difficult to use. Similarly, a voice-based system may perform differently depending on a person’s accent, speech pattern, hearing ability, or speech disability.<\/p>\n\n\n\n

Older adults, people with disabilities, family caregivers, and direct care workers can provide valuable insight during technology selection, testing and ongoing review.<\/p>\n\n\n\n

Their experiences may reveal barriers that are not obvious during a demonstration.<\/p>\n\n\n\n

Choice matters, too. Some people may prefer a face-to-face conversation, a telephone call, written communication, or another approach. When technology does not meet an individual’s needs, an appropriate alternative should remain available.<\/p>\n\n\n\n

Question 5: Does It Create More Human Attention?<\/h2>\n\n\n\n

The goal of care technology should be to support people, not simply to increase the number of tasks completed.<\/p>\n\n\n\n

If an AI tool reduces repetitive administrative work, a caregiver may have more time to listen, observe, communicate and respond to the person receiving care. On the other hand, a system that requires frequent corrections, monitoring, or duplicate documentation could increase rather than reduce workload.<\/p>\n\n\n\n

For that reason, organizations should look beyond efficiency metrics.<\/p>\n\n\n\n

Questions about human-centered AI in direct care<\/strong> should include whether a technology affects job quality, trust, continuity, communication and the amount of time available for meaningful relationships.<\/p>\n\n\n\n

\"Technology<\/figure>\n\n\n\n

Keep Accountability Human<\/h2>\n\n\n\n

AI can support a workflow, but it does not carry professional responsibility.<\/p>\n\n\n\n

Organizations remain responsible for privacy, safety, accessibility, supervision and applicable requirements. Direct care workers remain responsible for working within their roles and raising concerns when something does not appear right.<\/p>\n\n\n\n

The importance of human judgment becomes especially clear in home care. A familiar routine, a subtle change in behavior, a person’s communication preference, or a family caregiver’s observation may provide context that an automated system cannot fully understand.<\/p>\n\n\n\n

Technology should support that human knowledge rather than replace it.<\/p>\n\n\n\n

What Human-Centered AI Could Mean for Home Care<\/h2>\n\n\n\n

For home care organizations, responsible technology use may involve starting small, clearly defining the problem, testing a tool with the people who will actually use it and regularly reviewing its effects.<\/p>\n\n\n\n

Families can participate in that process by asking how technology is being used and how it affects the person’s care.<\/p>\n\n\n\n

Some helpful questions include:<\/strong><\/p>\n\n\n\n