Should We Welcome AI, Fear It, Or Both?

Few things have come into our lives and fundamentally changed how we work more quickly than artificial intelligence (AI). From writing to software engineering to app development and beyond, AI’s rapidly growing capabilities have many job sectors rethinking what their futures look like—includinghealth care.

A study¹ published in BMJ Global Health1explores the potential wide-reaching consequences of AI in multiple areas of human life, health care included. “AI holds the potential to revolutionize healthcare by improving diagnostics, helping develop new treatments, supporting providers and extending healthcare beyond the health facility and to more people,” the study authors wrote.

And while it’s not hard to decipher how the pharmacy field could be improved by AI—new treatments and even more accessibility thanks to AI will help pharmacists serve larger populations—the researchers urged serious caution as AI is incorporated into daily life. It is also important to explore the potential dangers of AI—some of which have gone from potential to actual dangers—in the health care space.

The arrival of ChatGPTisn’t the first time automation has appeared in health care or pharmacy; adherence packaging, pharmacy inventory, and medication ordering have utilized automation in one way or another, well before theAI boom began. However, none of those uses have posed “threats to human health and well-being via social, political, economic and security-related determinants of health” like the new wave of AI does, according to the study authors.

Problems of objectivity and biases have already begun to arise in the health care field from AI technology, as one example highlighted by the authors shows that is already does not.

“One example of harm accentuated by incomplete or biased data was the development of an AI-driven pulse oximeter that overestimated blood oxygen levels in patients with darker skin, resulting in the undertreatment of their hypoxia.” Biases like these already appearing in AI show just how flawed an objective technology can be. “Facial recognition systems have also been shown to be more likely to misclassify gender in subjects who are darker-skinned. It has also been shown that populations who are subject to discrimination are under-represented in datasets underlying AI solutions and may thus be denied the full benefits of AI in healthcare.”

The researchers of the study are not discrediting AI—rather, they suggest taking a step back, as we have already begun to see potential problems in health equity and practices due to AI. And in the early stages of AI incorporation, fixing those issues now may be paramount, they say.

“Perhaps the most important thing is to simply raise the alarm about the risks and threats posed by AI,” the researchers concluded in regard to the health care community, “And to make the argument that speed and seriousness are essential if we are to avoid the various harmful and potentially catastrophic consequences of AI-enhanced technologies being developed and used without adequate safeguards and regulation.” 

Reference

1. Federspiel  F, Mitchell R, Asokan A, et al. Threats by artificial intelligence to human health and human existence. BMJ Glob Health 2023;8:e010435. doi:10.1136/ bmjgh-2022-010435