Can AI Explain Obesity Better Than Humans?

Machine LearningMore than a few writers have been worrying lately about a new artificial intelligence app, ChatGPT. In the Guardian, Alex Hern supposed that it will put professors and journalists out of their jobs. So we wondered. Could this implementation of AI demonstrate an understanding of obesity?

The obvious answer was to give it a try.

What Is Obesity?

So we asked. Is obesity a legitimate medical diagnosis or a stigmatizing label for someone living in a larger body? This was the response:

“Obesity is a legitimate medical diagnosis. It is defined as a condition in which a person has an excess amount of body fat, to the point that it may have an adverse effect on their health. Obesity is diagnosed using a measure called the body mass index (BMI), which is calculated using a person’s weight and height. A BMI of 30 or higher is generally considered to indicate obesity.

“Obesity can have serious consequences for a person’s health, including an increased risk of developing conditions such as type 2 diabetes, heart disease, and certain types of cancer. For these reasons, it is important for people who are obese to receive proper medical care and support in order to improve their health and reduce their risk of developing these conditions.

“While the term ‘obesity’ may sometimes be used in a stigmatizing way, it is important to remember that obesity is a medical condition and not a personal failing. It is important to treat people living in larger bodies with respect and understanding, and to recognize that obesity is a complex issue that may be influenced by a variety of factors, including genetics, environment, and lifestyle.”

A Passable Understanding

The AI response to this basic question about obesity was surprisingly good, though not perfect. We would quibble with the suggestion that BMI alone can serve to diagnose it. We also noticed that in one instance, it wrote about “people who are obese.”

Person-First Language?

The obvious follow-up question, then, was about labeling a person as “obese.” When a person has an excess amount of body fat that harms their health, is it more accurate to say that person is obese or that they have obesity? The answer was deft:

“It is more accurate to say that a person has obesity if they have an excess amount of body fat that harms their health. Obesity is a medical condition that is diagnosed based on a measure called the body mass index (BMI), which is calculated using a person’s weight and height. A BMI of 30 or higher is generally considered to indicate obesity.

“Obesity is not a personal characteristic or trait, and it is not something that a person is. Rather, it is a medical condition that a person has. Therefore, it is more accurate to say that a person has obesity rather than saying that they are obese.

“It is important to use language that is sensitive and respectful when discussing obesity and other medical conditions. Using language that emphasizes the person rather than the condition can help to reduce stigma and promote understanding and empathy.”

ChatGPT even offered a reasonable explanation for Health at Every Size and its limitations. “A holistic approach to health and wellness” that has “faced criticism and controversy.”

Chill

Our little glimpse of what ChatGPT can do impresses us. Maybe AI can offer prose that reflects a better understanding of obesity than many humans can. It seems more like an interesting tool than an ominous threat.

For further perspective, click here, here, here, and here.

Machine Learning, illustration by David S. Soriano, licensed under CC BY-SA 4.0

Subscribe by email to follow the accumulating evidence and observations that shape our view of health, obesity, and policy.


 

December 18, 2022