Insights |

Generative AI for Healthcare Providers and Payers: Use Cases and Risks

by Chad Holmes

Generative AI is poised to change society as we know it. From creating new antibodies and passing bar exams to winning art competitions and diagnosing patient symptoms, generative AI is revolutionizing the way humans interact with technology.

In healthcare, generative AI has the potential to accelerate the transformation of an entire industry. AI can demystify complex patient diagnoses, provide timely and insightful information while reducing administrative burden for physicians, and improve the accuracy and speed of insurance claims.

By generating new healthcare data solutions, along with harvesting insight from volumes of historical data, AI can help patients and providers make better decisions, reduce costs, improve member experiences, and lead to better overall.

An AI-generated image of a sprialing rope reminiscent of DNA

Generative AI vs.traditional AI use cases in healthcare

To identify new potential use cases for healthcare, it’s important to understand the benefits of generative AI in comparison to traditional AI use cases, such as chatbots and virtual assistants. These include:

  • High-fidelity intent recognition: Using deep neural network algorithms and domain-specific ontologies to detect intent more accurately in long sequences, versus keyword matching or simple algorithmic models
  • Sentiment analysis capabilities: The ability to understand human emotions and moods
  • Flexible, free-flowing dialogue: Using natural language processing (NLP) and natural language understanding (NLU) to understand context and speak in coherent, human-like sentences
  • Inferring and self-directed transactional capabilities: Bots being able to handle variance, as opposed to following decision trees and scripts
  • Self-attention mechanism: Introducing the concept of memory and enabling a model to weigh the importance of different elements in an input sequence

Given the massive amounts of data the healthcare system handles, including claims data, population data, clinical data, health records, and images, the industry is ideal for automated processes. But how can generative AI go further than the existing AI use cases?

Impact of generative AI on healthcare providers

One of the most promising applications of generative AI in healthcare is in the area of medical diagnosis. By analyzing large amounts of medical data, generative AI can identify patterns and trends that would be difficult or impossible for humans to detect. That helps providers diagnose diseases earlier and more accurately, leading to better outcomes for patients.

Generative AI can also be used to create personalized treatment plans for patients. By taking into account a patient's medical history, lifestyle, and preferences, AI can generate treatment plans that are more likely to be effective. That means targeted care management, improved outcomes, and a reduced cost of care for the patient and their insurer.

In addition to diagnosis and treatment, generative AI can also be used to improve operational efficiency. For example, providers can use AI to automate tasks such as scheduling appointments, managing patient records, and generating billing reports in ways that go beyond traditional robotic process automation (RPA) by including prioritization needs and other factors.

Here are a few potential provider use cases that use AI to deliver deeper insights and efficacy:

1. Improved patient care

Generative AI can help healthcare providers:

  • Identify diseases earlier and more accurately
  • Increase speed and accuracy of basic triage
  • Create personalized treatment plans
  • Provide virtual health assistance, including medication reminders
  • Improve customer care with personalized, automated communication between patients and providers, such as intelligent chatbots for preliminary triage diagnostics

2. Improved efficiencies and reduced costs:

Generative AI can amplify traditional AI-enabled uses cases to furthe reduce healthcare costs by:

  • Automating tasks such as scheduling appointments, auto-populating intake forms, auto-procedure coding (ICD-10, HCPCS), and managing patient records
  • Enabling medical simulation for more effective training and research
  • Automating revenue cycle automation functions such as claims editing, prior authorization and auto-form filling, and denial appeals
  • Predicting and monitoring maintenance needs
  • Using enhanced telemedicine and surgical robots

3. Better decision-making:

Generative AI can supply providers with access to information and insights in areas like:

  • Diagnostics and screening – using deep learning-based AI algorithms to automatically detect complex anomalous patterns in images within seconds, particularly in radiology
  • Clinical trial support – identifying trends in large datasets by quickly mining accurate, relevant, evidence-based information that has been curated by medical professionals
  • Evidence-based recommendations – providing relevant findings from patient records, as well as outcome predictions
  • Value-based care – identifying and accelerating delivery of offerings
  • Digital twins – creating a digital representation of a person’s physical attributes that keeps a record of medical history accessible to healthcare providers through the Electronic Health Record (EHR)

A CAT scan showing many cross-sections of a brain

Impact of generative AI on healthcare payers

Generative AI makes it easier than ever for insurance payers to uncover care patterns and insights. These insights, combined with automation opportunities, will lead to vast improvements in payer operations, ultimately benefiting the entire healthcare ecosystem by reducing costs and helping make healthcare more accessible and affordable for everyone.

Here are a few potential payer use cases that yield deeper insights and greater efficacy:

1. Better decision-making

AI can provide payers access to more timely and richer information and insights in areas like:

2. Improved efficiencies and reduced costs:

Generative AI can complement traditional AI to speed up and improve performance of operational and clinical workflows and processes with the following:

  • Automated/intelligent claims review and edits
  • Automated prior authorizations
  • Auto-prediction of referral requirements and analysis of provider referral patterns
  • Replacement of predominately manual tasks like member verification, provider credentialing, and accounts payable
  • Customer care using chat, predictive analytics, and automated suggestions


Risks of generative AI in healthcare

As the technical barrier to entry for creating and deploying generative AI systems has lowered dramatically, the ethical issues and risks around AI have become more apparent. Some common risks associated with its use include:

  1. Inherent bias: Generative AI models can be biased, which can lead to inaccurate or unfair results. This is a particular concern when the models are trained on data that is itself biased.
  2. Hallucinations:” Generative AI can present information as factual when it’s actually not true.
  3. Copyright issues: Thus far, the U.S. Copyright Office has held that there is no copyright protection for works created by non-humans, including machines. Therefore, content or an image created by an AI model can’t be copyrighted, leading to unclear infringement scenarios.
  4. Misinformation/fake news: Generative AI models could be misused by bad actors to create fake medical information or to spread misinformation. This could have a negative impact on patient care and public health.
  5. Security: Generative AI models can be vulnerable to hacking and other security threats. This could lead to theft of sensitive patient data or the use of models to generate fake medical information.
  6. Emergent properties: In such early days of generative AI, some people have concerns about unauthorized decision-making and “self”-taught skills on the part of AI bots.

Especially in an industry as concerned with privacy, security, and compliance as healthcare is, it’s important to be aware of these risks so they can be mitigated and monitored. This can be done by carefully designing and testing generative AI models, using them in a responsible way, and educating healthcare providers about the risks.

An AI-generated image of a brain with a digital 3D-model effect

The future of generative AI

There's little debate that generative AI is a powerful tool that has the potential to revolutionize healthcare. However, choosing use cases carefully and undergoing a secure and robust implementation is imperative as we develop a more comprehensive understanding of the legal and security picture surrounding generative AI in healthcare. By judiciously designing and testing AI models, using them in a responsible way, and educating all involved about the risks, we can better ensure that generative AI is used for good: improving decisions, lowering the cost of healthcare, and achieving better patient outcomes.


About the Author:

Chad Holmes is a Senior Managing Partner with Launch Consulting and the Managing Director for Launch’s Healthcare Sector. His focus areas include human experience design, data transformation, intelligent operations, and transformational change leadership. With expertise across healthcare operations, data science, and digital transformation capabilities, Launch is ready to help organizations navigate successful AI adoption for the long term.

Back to top

More from
Latest news

Discover latest posts from the NSIDE team.

Recent posts
This is some text inside of a div block.

Launch Consulting Logo