Search Engines and Generative Artificial Intelligence Integration: Public Health Risks and Recommendations to Safeguard Consumers Online

AimArtificial intelligence (AI) chatbots such as Chat Generative Pretrained Transformer‐4 (ChatGPT‐4) have made significant strides in generating human‐like responses. Trained on an extensive corpus of medical literature, ChatGPT‐4 has the potential to augment patient education materials. These chatbots may be beneficial to populations considering a diagnosis of colorectal cancer (CRC). However, the accuracy and quality of patient education materials are crucial for informed decision‐making. Given workforce demands impacting holistic care, AI chatbots can bridge gaps in CRC information, reaching wider demographics and crossing language barriers. However, rigorous evaluation is essential to ensure accuracy, quality and readability. Therefore, this study aims to evaluate the efficacy, quality and readability of answers generated by ChatGPT‐4 on CRC, utilizing patient‐style question prompts.MethodTo evaluate ChatGPT‐4, eight CRC‐related questions were derived using peer‐reviewed literature and Google Trends. Eight colorectal surgeons evaluated AI responses for accuracy, safety, appropriateness, actionability and effectiveness. Quality was assessed using validated tools: the Patient Education Materials Assessment Tool (PEMAT‐AI), modified DISCERN (DISCERN‐AI) and Global Quality Score (GQS). A number of readability assessments were measured including Flesch Reading Ease (FRE) and the Gunning Fog Index (GFI).ResultsThe responses were generally accurate (median 4.00), safe (4.25), appropriate (4.00), actionable (4.00) and effective (4.00). Quality assessments rated PEMAT‐AI as ‘very good’ (71.43), DISCERN‐AI as ‘fair’ (12.00) and GQS as ‘high’ (4.00). Readability scores indicated difficulty (FRE 47.00, GFI 12.40), suggesting a higher educational level was required.ConclusionThis study concludes that ChatGPT‐4 is capable of providing safe but nonspecific medical information, suggesting its potential as a patient education aid. However, enhancements in readability through contextual prompting and fine‐tuning techniques are required before considering implementation into clinical practice.

show abstract

Large Language Models Can Enable Inductive Thematic Analysis of a Social Media Corpus in a Single Prompt: Human Validation Study

Deiner,

Honcharov,

et al. 2024

JMIR Infodemiology

View full text Add to dashboard Cite

Background: Manually analyzing public health-related content from social media provides valuable insights into beliefs, attitudes, and behaviors of individuals, shedding light on trends and patterns that can inform public understanding, policy decisions, targeted interventions, and communication strategies. Unfortunately, the time and effort needed from well-trained human subject matter experts makes extensive manual social media listening unfeasible. Generative large language models (LLMs) have potential for not only summarizing but also interpreting large amounts of text. It is not clear to what extent LLMs can analyze a large set of social media posts at once to glean subtleties of health-related meaning and reasonably report on healthrelated themes.Objective: Assess feasibility of using LLMs for topic model selection or inductive thematic analysis of large contents of social media posts: Can LLMs conduct topic model selection and inductive thematic analysis as effectively as humans did in a prior manual study, or at least reasonably, as judged by subject matter experts? Methods: We asked the same research question and used the same set of social media content for both the LLM selection of relevant topics and the LLM analysis of themes, as was conducted manually in a prior published study about vaccine rhetoric. We compared the results from the prior manual human analyses to results from analyses by the LLMs GPT4-32K, Claudeinstant-100K, and Claude-2-100K. We also assessed if multiple LLMs had equivalent ability and assessed consistency of repeated analysis from each single LLM.Results: Overall, all three LLMs could assess the large corpus of social media posts and summarize content. LLMs generally gave high rankings to the topics chosen previously by humans as most relevant. We reject a null hypothesis (P<0.001, overall comparison) and conclude that these LLMs are more likely to include the human-rated top 5 content areas in their own top rankings than would occur by chance. Regarding theme identification, LLMs identified several themes similar to those identified by humans, with very low hallucination rates. Variability occurred between LLMs and between test runs of an individual LLM. Despite not consistently matching the human-generated themes, subject matter experts found themes generated by the LLMs were still reasonable and relevant.Conclusions: LLMs can effectively and efficiently process large social media health-related datasets. LLMs can extract themes from such data that human subject matter experts deem reasonable. However, we were unable to show that the LLMs we tested can replicate the depth of analysis from human subject matter experts by consistently extracting the same themes from the same JMIR Preprints Deiner et al

show abstract

Search Engines and Generative Artificial Intelligence Integration: Public Health Risks and Recommendations to Safeguard Consumers Online

Cited by 6 publications

References 33 publications

The Evolution of Information Seeking Practices and Technologies and the Future of the Library

The Evolution of Information Seeking Practices and Technologies and the Future of the Library

ChatGPT as a patient education tool in colorectal cancer—An in‐depth assessment of efficacy, quality and readability

Large Language Models Can Enable Inductive Thematic Analysis of a Social Media Corpus in a Single Prompt: Human Validation Study

Contact Info

Product

Resources

About