Can users trust the cancer-related information provided by ChatGPT, the artificial intelligence chatbot created by OpenAI? Recently, this question has been a concern amongst tech experts especially after reports of biases and harmful content generated by bad users surfaced. To address the issue, a new study was made to determine the accuracy of ChatGPT’s cancer-related answers.
The study was based on the National Cancer Institute’s (NCI) list of cancer myths and misconceptions. Headed by Skyler Johnson, a professor from the University of Utah, researchers determined that ChatGPT provided cancer information with an average accuracy of 97%. Nonetheless, Johnson himself noted that this artificial intelligence chatbot can still lead to bad decisions if its answers are misunderstood or misinterpreted.
ChatGPT’s cancer-related answers can sometimes be unclear, vague, or even indirect. Johnson recommends that cancer patients and their caregivers should be well-informed and cautious about the chatbot’s responses in order for them to properly understand the online information available to them.The newly released report was posted on the National Cancer Institute Cancer Spectrum website for further review.
Meanwhile, last month, New York City public schools lifted the ban or limitation of ChatGPT in classrooms. The ChatGPT iOS was also expanded to other countries such as Brazil and Canada.
ChatGPT is an artificial intelligence chatbot created by OpenAI, a non-profit AI research organization. It was an initiative to create AI models that can perform tasks we thought were only limited to humans. With the help of machine learning and natural language processing, this artificial intelligence chatbot can engage advanced conversations with people. It works by having people type in prompts for it to answer.
Skyler Johnson is a professor from the Computer Science & Engineering department of the University of Utah. He specializes in research on natural language processing and user interfaces. Johnson was also a part of the research team who conducted the study on ChatGPT’s accuracy in providing cancer-related information.
