Mobile Users Bet on AI for Personal Health Queries

Microsoft research indicates that mobile users increasingly rely on AI assistants for personal health and emotional support questions. This trend highlights a growing demand for digital health tools, which benefits specialized platforms such as Yana and accelerates innovation across the mental health technology sectors.

People who ask health questions through mobile devices tend to raise more personal and urgent concerns than those using desktop computers, according to new data from Microsoft analyzing more than 500,000 conversations with its Copilot AI assistant in January 2026. The findings indicate that device choice influences how users interact with AI for health-related information and support.

Health inquiries represent one of the most frequent uses of Microsoft’s Copilot platform. According to Dominic King, Vice President, Microsoft AI, demand for health guidance continues to outpace access to medical professionals in many contexts. “It is such a substantial problem and such a growing problem,” King said. “This feels like the easiest path for now for people to get their expertise and support.”

The company says the analysis illustrates how generative AI (GenAI) is increasingly used to fill gaps in access to medical information, particularly outside standard clinic hours.

Mobile devices drive more personal health conversations

Microsoft researchers analyzed health and well-being conversations held with Copilot during Jan. 2026. The dataset included more than half a million interactions that were de-identified and processed through automated systems designed to extract topics and user intent without human review.

The results suggest that smartphones have become the primary channel for personal health discussions with AI systems. Users on mobile devices were more likely to ask about symptoms affecting themselves or a family member. Emotional well-being conversations were also significantly more common on mobile platforms.

By contrast, desktop interactions skewed toward research-oriented questions. Academic research, medical literature inquiries and other informational requests appeared three times more often on desktop computers than on mobile devices.

Researchers found that questions related to symptoms and condition management occur roughly twice as often on phones as on computers. Emotional well-being conversations were about 75% more frequent on mobile devices.

Overall, approximately 40% of health-related conversations focused on general medical information such as symptoms, conditions and treatments. Nearly 11% involved detailed questions about interpreting symptoms or medical test results. About 6% of queries related to navigating health care systems, including insurance coverage, provider access or medical benefits.

Other conversations addressed topics such as fitness coaching, academic research, emotional well-being and medical paperwork.

Microsoft’s analysis also indicates that device usage reflects the context in which questions arise. Mobile interactions often occur in real time when users face an immediate concern, while desktop use tends to align with structured research or professional work.

AI use increases when traditional care access is limited

The timing of Copilot health questions provides additional insight into how people rely on AI for information. The study found that health queries evolve throughout the day, with more personal or emotionally focused questions appearing later in the evening.

Emotional well-being questions accounted for about 3.4% of health-related conversations during the morning and daytime hours. That share rose to 4.3% in the evening and to 5.2% overnight.

Researchers also observed an increase in symptom-related questions during nighttime hours, when users may have fewer options to consult clinicians, pharmacists or family members.

The findings suggest that AI assistants are often used when traditional sources of health information are unavailable. GenAI systems can provide explanations, guide users through follow-up questions and help interpret health information before users decide whether to seek medical care.

Across Microsoft’s consumer platforms, including Bing and Copilot, the company says it handles more than 50 million health-related questions each day.

As healthcare systems face rising demand and workforce shortages in many regions, technology companies have begun positioning AI assistants as tools that can provide preliminary information and guide people toward clinical services.

Specialized AI platforms focused on mental health are also emerging within this landscape. One example is Yana, an AI-powered emotional companion designed to provide accessible mental health support through structured conversations based on cognitive behavioral therapy (CBT) techniques.

According to the company, Yana was originally built as a chatbot that guided users through decision-tree pathways designed to help them identify and challenge negative thoughts. 

With the integration of AI, the system can now adapt its responses more dynamically, recognizing patterns in user conversations and offering more personalized guidance in real time. This allows the platform to move beyond predefined pathways. Instead, the system responds to users’ emotional context and helps them challenge cognitive distortions.

Specialized mental health AI platforms expand

While general-purpose AI assistants such as Copilot or ChatGPT increasingly receive questions related to mental health and emotional well-being, specialized platforms argue that purpose-built tools offer advantages when dealing with sensitive topics.

Andrea Campos, CEO and Founder, Yana says that mental health conversations require structured guidance rather than open-ended exchanges. 

“A therapist is specifically trained to guide conversations in a way that facilitates meaningful progress,” Campos explained. “In therapy, the ability to be guided through the process is essential. This is why Yana is designed to manage and facilitate such conversations, even when dealing with sensitive topics.”

According to Campos, general-purpose AI systems are not designed specifically for therapy, which can create limitations when users seek psychological guidance. Platforms built specifically for mental health aim to incorporate safety frameworks, therapeutic structures and purpose-driven conversational design.

Since its launch, Yana has recorded approximately 16 million downloads globally, with about 12 million registered users. Around 30% of monthly downloads come from returning users who previously created accounts and return to the platform when facing new challenges.

Yana user demographics have evolved as mental health technology adoption expands. Previously, teenagers represented the majority of users, but today about 70% of active users are adults aged 18 or older. Developers view this shift as an indication that barriers to accessing mental health support tools may be decreasing.

Despite technological changes, the underlying issues users discuss remain relatively consistent. According to Campos the most common concern users bring to the platform is low self-esteem, followed by challenges related to stress, work pressures, relationships and family dynamics.

The company attributes the prominence of self-esteem concerns partly to the growing influence of social media, where constant comparison and public scrutiny can intensify feelings of inadequacy.

Looking ahead, developers say AI could play an expanding role in addressing the global shortage of mental health services. Many healthcare systems face limited numbers of therapists relative to demand, while cost and access barriers prevent many individuals from seeking professional help.

In this context, AI-based tools may function as complementary support rather than replacements for clinical care. Yana’s developers envision a hybrid model in which therapists conduct periodic sessions while AI systems provide continuous support between appointments.

Such an approach could help identify behavioral patterns earlier, prepare patients for more productive clinical conversations and potentially shorten treatment timelines by providing ongoing engagement.

For Microsoft, the Copilot research underscores the importance of accuracy and reliability when AI systems respond to health questions. The company says it is developing models designed to provide stronger reasoning and richer clinical context in responses, helping users understand health information and prepare for discussions with clinicians.