Embracing the Power of AI and LLM in Health Care: The WHO Perspective
Artificial intelligence (AI) and large language model tools (LLMs) such as ChatGPT, Bert, and Bard have gained notable traction in healthcare since their introduction in September 2022. In a recent press release, the World Health Organization (WHO) expressed its enthusiasm toward the potential of these technologies, primarily when they are used appropriately to promote human welfare, security, and public health. The WHO caution follows on the heels of a similar optimistic but cautionary note for health care AI by the US White House.
The rise of LLM platforms has seen rapid expansion, with their features designed to emulate human comprehension and communication being increasingly embraced. Their growing experimental use in health-related purposes brings excitement for their potential to support health needs, as reported by the WHO in a recent press release.
The WHO emphasizes that when used correctly, LLMs can benefit many stakeholders, including healthcare professionals, patients, and researchers. However, it also urges due diligence in managing the associated risks and calls for a thorough examination to better health information access, enhance diagnostic capacities, and reduce inequities.
The Risks of Health Care AI and WHO’s Call for Caution
Concerns have been raised that the general precaution for new technologies may differ from LLMs. The WHO health care AI advocates for adherence to key principles such as transparency, inclusion, public engagement, expert supervision, and rigorous evaluation. Unmeasured adoption of these systems could lead to mistakes by healthcare providers, potentially harming patients and eroding trust in AI.
WHO identified several risks for health care AI, including the following:
Biased data used to train AI could generate misleading or false information that might endanger health, equity, and inclusiveness.
Authoritative responses generated by LLM platforms might contain inaccuracies, especially with health-related responses.
Misuse of sensitive health data provided by users could be used for training without consent.
The potential misuse of LLMs creates convincing disinformation that is difficult to distinguish from reliable health content.
WHO strongly recommends addressing these concerns before these technologies are widely used in routine healthcare and medicine. It highlights the need for clear evidence of benefits, particularly for healthcare providers and policymakers.
AI and LLMs in Action: Easing Burnout Due to Office-Related Tasks
The use of health care AI and LLMs has the potential to ease burnout due to office-related tasks. Various applications have been identified, from creating a cohesive patient journey to providing real-time, intelligent support for agents and addressing signs of burnout.
Streamlining the Patient Journey with AI
Many healthcare organizations are investing in digital technologies to enhance the patient experience. However, patients often need help navigating their care journey due to disconnected engagement tools and a lack of support. Integration is necessary for critical information gaps, improving patient experiences, and operational inefficiencies.
A solution to these challenges lies in deploying technologies that work together to deliver a seamless patient journey. Health care AI-powered cloud-based contact centers can help eliminate the fragmented nature of the patient experience. They can make omnichannel patient communication easier, harmonize the patient journey, and reduce friction