The use of generative artificial intelligence (AI) in healthcare is growing quickly across the United States. Both big health systems and smaller medical offices are starting to use AI tools to make administration easier, improve clinical work, and increase patient engagement. Many healthcare leaders see possible benefits in using generative AI. Still, there are some challenges. These mainly involve following regulations, managing risks, and fixing gaps in their own skills and systems. For practice managers, owners, and IT staff in the U.S., knowing these barriers is important for using AI successfully.
This article talks about key challenges healthcare groups in the U.S. face when using generative AI. It also shows how important it is to build partnerships, have good risk management, and automate workflows.
Recent studies show that many healthcare leaders in the U.S. are becoming more interested in generative AI. A McKinsey survey from late 2024 found that 85% of healthcare leaders are either looking into or already using generative AI. This includes executives from large organizations with over $10 billion in revenue and smaller, specialized healthcare groups.
Most groups prefer to work with outside companies instead of building AI tools themselves. About 61% choose to partner with vendors to create custom AI solutions. Only 20% want to develop AI systems internally, and 19% use ready-made AI products.
The first ways AI is being used focus on tasks like scheduling, billing, answering calls, and making clinical work like documentation easier. Many say they have seen or expect good financial results from their AI projects. In fact, 64% confirm positive economic impact from AI.
Even with these early successes, there are still problems that slow down the smooth use and growth of AI in healthcare settings.
One main challenge in using generative AI in healthcare is dealing with complicated rules that apply to the U.S. medical system. Healthcare groups must follow federal laws like the Health Insurance Portability and Accountability Act (HIPAA). This law protects patient data privacy and security. They also must meet standards set by the Food and Drug Administration (FDA) for medical devices and clinical software.
Many AI tools that interact with patients or help make clinical decisions are in a regulatory area that is still changing. This creates uncertainty about how these AI tools are classified and checked by regulators. Not following rules can lead to big fines and lose trust from the public.
There are also worries about data privacy and bias in AI. AI needs lots of data, and if there are problems with how data is managed, patient information could be leaked or used wrongly. Experts like Baily Ramsey say that ethical rules and transparency are important to follow laws and make sure AI is used fairly.
Healthcare organizations often find it hard to balance new technology with rules because federal and state laws may differ and change slowly compared to fast AI development. This can cause delays or make groups use safer but less innovative AI methods.
Another big challenge for U.S. medical practices is a lack of skills and resources to use AI well. Many healthcare providers do not have enough technical knowledge, data systems, or project management ability for AI adoption.
One key issue is a shortage of AI engineers and data scientists inside organizations. Reports show filling these roles takes over two months on average because many companies want these experts. This shortage makes it hard for medical managers and IT people to keep AI tools working or change them for their needs without help from outside experts.
Older systems used in healthcare—like electronic health records (EHRs), billing, and customer management platforms—often do not work well with new AI tools. These compatibility problems cause delays and reduce how well AI can perform.
Because of this, many healthcare groups hire outside consultants and technology vendors who know about AI systems. McKinsey’s survey says 46% of groups look for partnerships with large cloud providers to fix these gaps and speed up AI use.
Healthcare leaders say managing risk and having clear governance is very important when using generative AI. Introducing AI that helps with patient care or clinical decisions needs careful oversight to keep it safe, reliable, and responsible.
Governance means setting rules, processes, and controls to check AI models all the time for following laws, accuracy, and bias. Because AI changes fast, governance must adapt to new ethical issues and updated rules.
Experts like Carlos Pardo Martin point out that success with AI takes more than buying technology. Organizations need proper systems to deliver benefits and lower risks. They need strong strategies and good execution along with governance so AI works well with clinical and admin goals without causing problems.
Healthcare providers must also watch AI models continuously. AI can work less well or act unpredictably over time, especially with new data. Keeping AI accurate and fair is important to avoid harming patient care or causing legal problems.
One useful way generative AI is used in healthcare is in front-office tasks like answering calls and handling patient interactions. Companies like Simbo AI focus on AI-powered phone automation designed for medical offices, clinics, and hospitals in the U.S.
AI can do routine jobs such as scheduling appointments, filling prescriptions, checking insurance, and answering patient questions. This helps front-desk staff by lowering their workload and reducing wait times. AI systems can handle many calls accurately and improve patient experience and productivity.
Automation with AI also helps with staff shortages common in healthcare. When AI takes over repetitive tasks, healthcare workers can focus on more complex things that need human judgment.
AI tools for workflow automation are made to work with old healthcare IT systems, fixing some problems caused by legacy software. This helps close the gap in skills and technology.
Baily Ramsey notes that success with AI automation depends on good planning. This includes checking current systems, preparing data carefully, choosing the right AI models, and testing results before full use. Ongoing checks keep automation effective and controlled.
Automation also indirectly helps with following rules by lowering human errors and making processes more consistent. For example, AI can keep accurate call logs and documentation, which is important for audits.
Because using generative AI can be difficult, building strategic partnerships is important for many U.S. healthcare organizations. Working with AI vendors, consultants, and big cloud providers helps healthcare groups use expert knowledge, powerful data tools, and scalable technology they might not have inside.
Partnerships give flexibility and allow organizations to customize AI for their specific clinical and front-office needs without making big internal teams. These partnerships also help deliver AI tools faster and avoid delays that happen when organizations try to build everything themselves.
McKinsey’s research shows that healthcare groups working with external partners usually get better returns on AI investments. Partners help with AI readiness checks, improving infrastructure, following regulations, and supporting systems after they launch.
Good data is key for successful generative AI. But many healthcare groups struggle with poor data quality, which causes AI models to give wrong or unreliable information. This can put patient care and management decisions at risk.
Old electronic systems often have incomplete or scattered data. Many providers don’t know if their data is “AI-ready,” meaning it’s well-organized, complete, and consistent enough to train and test AI models properly.
Experts say healthcare groups should check data readiness before using AI. This involves finding missing data, fixing errors, and making data consistent across sources.
Older systems also don’t work well with modern AI interfaces, making data exchange difficult. Many EHRs and billing systems were not built for AI technology, so upgrades or extra software may be needed.
Investing in upgrades like cloud services and better APIs helps solve these problems and supports ongoing AI projects.
Using generative AI successfully doesn’t depend only on technology. It needs strong leadership and management in healthcare organizations. Leaders like Jessica Lamb point out that groups with clear AI plans and internal skills usually do better at expanding AI use.
Leadership must:
Without strong leadership, AI projects often remain small pilots and fail to become a regular part of daily work or meet financial goals.
Generative AI brings chances to improve healthcare in the U.S., especially by making administration and clinical work more efficient. But there are challenges with rules, skills, data systems, and risk management that slow down progress.
Healthcare groups need to think carefully about these issues when using AI. Working with specialized vendors and consultants, investing in good data systems, and creating clear governance can help manage complexities.
Automation of workflows, like AI answering phones, is a practical way to reduce admin work now. By fixing gaps in talent, technology, and strategy, medical managers and IT staff can make AI use safer and more efficient for both providers and patients.
85 percent of surveyed healthcare leaders, including payers, health systems, and HST groups, reported they are exploring or have adopted generative AI capabilities.
61 percent of respondents indicated that partnerships with third-party vendors to develop customized solutions were their primary strategy for adopting generative AI.
Respondents identified improving administrative efficiency and clinical productivity as areas with the greatest potential for generative AI to create value.
64 percent of those who have implemented generative AI use cases reported that they anticipated or had already quantified positive ROI.
Early use cases focused on improving administrative efficiency, addressing IT/infrastructure gaps, and increasing clinical productivity.
Challenges include evolving regulations, risk compliance complexities, and internal capability gaps that may hinder effective adoption.
Partnerships with hyperscalers provide expertise in data management and help ensure successful implementations of generative AI solutions.
AI governance is crucial for managing risks associated with implementing generative AI, ensuring safe advancement of these technologies.
Successful implementation requires a value-driven strategy, strong delivery capabilities, and robust organizational management to achieve at-scale impacts.
The survey targeted 150 US healthcare leaders across various subsectors, gathering insights on generative AI adoption, implementation stages, and anticipated benefits through responses and follow-up interviews.
Simbo is using AI agents to automate all phone related workflows. Connect now to know more.
Schedule AI Agents Demo Now© Since 2019, Simbo AI.
