Three steps healthcare organizations can take to use generative AI responsibly

The healthcare lead at research and consulting giant Accenture lays out how to get proprietary data ready, establish the right controls, and harmonize people with the tech.

In an interview with Rich Birhanzel, Accenture's healthcare lead, valuable insights on the responsible implementation of generative AI in healthcare are shared. Here are the key takeaways:

Getting Proprietary Data Ready: Large language models powering generative AI have the potential to process vast data sets, enabling them to understand an organization's entire knowledge base. While off-the-shelf models are used initially, true value is unlocked when customizing models with proprietary data. However, this requires access to domain-specific organizational data, making data readiness a priority. In healthcare, addressing the challenge of collecting, refining, and securing data becomes crucial, especially due to sensitive personal information. Cloud-based enterprise data platforms are recommended to meet this challenge.

Establishing the Right Controls: As generative AI adoption accelerates, healthcare organizations must define responsible AI missions and principles. Transparent governance structures are needed to build confidence and trust in AI technologies. Starting at the design stage, controls are essential to assess the potential risks of generative AI. These principles need to be top-down and align with laws, regulations, and organizational policies. A risk management framework with regular reporting and monitoring ensures responsible AI adoption.

Harmonizing People with Technology: Human guidance remains vital for generative AI applications in healthcare. Processes need refinement to accommodate AI capabilities and elevate the role of human workers. Training programs are necessary to equip workers with the skills needed to leverage AI effectively. In healthcare areas where generative AI is promising, decomposing tasks and assessing AI's impact on each task (automation, augmentation, or unaffected) is crucial. The potential lies in reducing burdens like healthcare documentation, transforming work dynamics, and enabling humans to collaborate more effectively with AI.

In the next five years, generative AI holds transformative potential in healthcare. AI and foundation models are revolutionizing how information is accessed, patient needs are met, and healthcare organizations operate. Healthcare providers are optimistic about the capabilities of generative AI to enhance delivery, accessibility, patient experience, and outcomes. This pivotal moment signifies an era of enterprise intelligence and transformative changes in the healthcare landscape.

Read on healthcareitnews.com