Google has removed its AI-generated health summaries from search results due to concerns over the accuracy of medical information, particularly regarding liver function tests. An investigation uncovered that the AI tool was delivering misleading health data, prompting the company to take action against specific summaries that posed potential risks to users.
The feature, known as AI Overviews, is designed to provide quick answers to users’ queries by generating summaries using artificial intelligence. However, an investigation revealed significant inaccuracies in some health-related content, raising alarms among medical professionals. In particular, the AI provided incorrect details about what constitutes normal ranges for liver blood tests, which could mislead patients about their health status.
Experts highlighted that the AI’s summaries lacked essential context, failing to consider variables such as a patient’s nationality, sex, ethnicity, or age. These factors can significantly influence what is considered a normal test result. For example, the AI indicated ranges for liver function tests that could misguide individuals, potentially leading them to overlook serious health conditions.
In response to the findings, Google has removed the AI Overviews for searches related to “normal range for liver blood tests” and “normal range for liver function tests.” A company spokesperson stated that Google does not provide comments on individual content removals but emphasized ongoing efforts to improve the relevance and reliability of health-related information.
Sue Farrington, chair of the Patient Information Forum, expressed support for the removal of the misleading summaries. She noted that while this decision is a positive step, it highlights broader issues regarding the trustworthiness of health information provided by AI. Farrington pointed out that millions of people worldwide struggle to access reliable health information, underscoring the critical need for Google to direct users to credible resources.
The investigation also identified other problematic AI Overviews still active on the platform, including summaries concerning cancer and mental health. Medical experts have described some of these as fundamentally erroneous and potentially dangerous. When questioned about the continuation of these summaries, Google stated that they link to reputable sources and advise users on the importance of seeking expert medical advice.
While the removal of certain AI-generated health summaries marks a proactive step by Google, the company still faces challenges in ensuring that all health-related content meets high accuracy standards. With the reliance on AI growing in the health sector, the need for rigorous oversight and verification of information becomes increasingly critical.
In an era where misinformation can have severe consequences, the integrity of health information remains paramount. Google’s actions serve as a reminder of the potential pitfalls of relying on automated tools for medical advice and the importance of consulting qualified healthcare professionals for accurate information.
