Artificial intelligence has quietly become standard in American healthcare, with three-quarters of organizations already deploying the technology. A survey of 277 healthcare professionals found that 75% of their organizations have mostly or fully embraced AI, yet concerns about patient relationships and data security remain significant barriers to full adoption .

What Are Healthcare Workers Actually Using AI For Right Now?

The most common application isn't diagnosing diseases or predicting patient outcomes. Instead, it's handling the administrative burden that's been driving physician burnout for years. Documentation and note-taking assistance tops the list, with 55% of healthcare leaders identifying it as their primary AI use case .

Companies like Ambience, Abridge, and Suki have emerged to fill this gap by using ambient listening technology, which records conversations between patients and providers and automatically generates clinical notes. The impact has been measurable: analysis from the New England Journal of Medicine Catalyst found that these AI scribes have reduced documentation work by more than 15,700 hours annually, work that often extends into what providers call "pajama time" when they're working from home after hours .

Beyond documentation, healthcare organizations are deploying AI across several other functions:

• Administrative Tasks: 45% of organizations use AI for prior authorization, appointment scheduling, and other back-office functions that slow patient access to care
• Voice Agents and Chatbots: 30% have implemented AI-powered phone systems and chat interfaces to handle patient inquiries and scheduling
• Patient Communication: 24% use AI tools, including real-time language translation services, to help patients understand their care
• Clinical Decision Support: 16% employ AI for diagnostics and treatment recommendations to assist physicians
• Population Health Management: 19% use predictive analytics to identify high-risk patients before they develop complications
• Research and Clinical Trials: 12% leverage AI to predict drug efficacy by comparing internal predictions against previous trial results and patient data

Electronic health record (EHR) companies themselves are getting into the AI game. Major players like Epic and Athenahealth have built their own scribe tools directly into their systems, and partnerships like Athena's collaboration with Microsoft are creating integrated solutions that give providers multiple AI options within a single platform .

How Is AI Changing the Patient-Doctor Relationship?

The shift toward AI documentation has produced an unexpected benefit: better patient interactions. When providers aren't typing notes during appointments, they can actually look at their patients and listen. Matthew Huddle, managing director and partner at consulting firm BCG's healthcare practice, noted that ambient scribes have been "a huge satisfier for staff" and are helping decrease burnout and turnover .

"It's also been a satisfier for the patients, too, because suddenly you're actually having a conversation with the physician who is looking at you and giving you their full attention versus mostly spending the time typing on a computer," said Matthew Huddle.

Matthew Huddle, Managing Director and Partner, BCG Healthcare Practice

However, this optimism isn't universal. In the survey, 38% of healthcare workers expressed concern about a loss of the "human touch" in care . These concerns reflect a broader anxiety about technology replacing human judgment and empathy in medicine.

Aimee Cardwell, chief information officer and chief information security officer in residence at data privacy company Transcend, suggested that comfort with AI-driven interactions may grow over time. "Eventually, we're going to get to the point where people are much more comfortable talking to machines than they are talking to people, and they feel like they can multitask better than trying to call somebody in a call center," Cardwell explained .

What Are the Biggest Concerns Holding Back AI Adoption?

Beyond worries about losing human connection, healthcare workers flagged two critical security issues. Nearly half of respondents reported concerns about cybersecurity risks, and approximately one-third worried about patient privacy violations . These concerns are not unfounded, as healthcare systems have become increasingly attractive targets for cyberattacks, and patient data breaches can expose sensitive medical and financial information.

The challenge for healthcare organizations is balancing the efficiency gains and cost savings that AI offers with the legitimate need to protect patient privacy and maintain the human elements of care that patients value.

How Might AI and Telehealth Work Together?

So far, AI and telehealth have operated largely in separate lanes. Only 11% of surveyed healthcare organizations said they weren't using telehealth, with 63% using it primarily for messaging and portal communication, 43% for behavioral health services, and 26% for remote monitoring .

But Huddle predicted convergence ahead. Some hospitals are already experimenting with AI triage or symptom checker tools that recommend patients schedule a telehealth visit if appropriate. "I think, as more of that automation and AI front door is added, you may see an increased utilization of virtual care," Huddle stated .

But Huddle

This combination could address one of healthcare's persistent challenges: care deserts and provider burnout. By automating routine administrative tasks and initial patient screening, AI could free up providers to focus on complex cases while telehealth extends their reach to patients in underserved areas.

What's the Financial Impact of Healthcare AI?

From a business perspective, AI tools focused on revenue cycle management, such as automating insurance prior authorization and claims processing, show the highest return on investment . These tools have "vastly increased the effectiveness of revenue cycle management teams and has led to increased collections," according to Huddle .

The broader economic potential is staggering. A Morgan Stanley analysis found that AI tools could save the healthcare system between $100 billion and $600 billion by 2050 just through improvements in drug development and clinical trial efficiency .

Despite these gains, the survey results suggest that healthcare organizations are still in the early stages of AI integration. The technology is solving immediate pain points like documentation burden and administrative inefficiency, but questions about privacy, security, and the role of human judgment in medicine remain unresolved. As AI becomes more embedded in healthcare workflows, these concerns will likely shape how the technology evolves and how patients experience their care.