-
FEATURED CONTENT
- 'Dr. ChatGPT' pays a visit to Beth Israel Deaconess. At a recent teaching event at Boston's Beth Israel Deaconess Medical Center, the chatbot GPT-4 worked alongside medical students to crack a real-life medical mystery: a 39-year-old woman who with a sore left knee and a 102-degree fever. The diagnosis? Lyme disease, a result reached both by the students and the AI. Still, one physician at the hospital warned, “[GPT-4 is] a great thought partner, but it doesn't replace deep mental expertise.”
- Related: A previously released study in JAMA found that GPT-4 performed well, if imperfectly, in diagnosing complex clinical case studies.
- Why OpenAI shuttered its AI detection tool. The short version: Their so-called “AI Classifier” just didn’t work. The tool, which aimed to distinguish between human and AI-written text, correctly identified only 26% of AI-generated content, while also mislabeling 9% of human-written content.
- Related : The White House and top AI firms last week announced voluntary commitments that included creating “mechanisms to ensure that users know when content is AI generated.” For now, though, there’s just no way to know whether a human or a machine wrote these words.
- America's top colleges, as determined by ChatGPT (in 10 minutes of work). Wharton's Ethan Mollick shared this transcript of a ChatGPT conversation in which he shared a raw dataset from the U.S. Department of Education and said, essentially, “Rank these colleges.” ChatGPT's Code Interpreter then extracted and interpreted the files, invented a ranking methodology, crunched the data, and announced its top 20 schools, complete with data visualizations. It's an impressive display of ChatGPT's increasingly sophisticated and autonomous coding abilities.
- Gamers fooled an AI news site into covering gobbledygook. Did you hear about “Glorbo,” the much-hyped new feature in World of Warcraft? Probably not … because Glorbo doesn't exist. It's nonsense invented by a Reddit community after they discovered an (apparently) AI-written website was turning their conversations into “news” articles. The ruse worked, revealing a new pathway for misinformation to spread in an AI-driven world.
- Your field guide to generative AI in healthcare. Explore this just-published roundup of healthcare use cases for generative AI from Advisory Board's Ty Aderhold. It's helpfully divided into near-term opportunities (such as automated documentation and coding), “next-level” use cases (including developing care management plans), and “moonshots” (could AI serve as a general diagnosis engine?)
- Related: To dive deeper into generative AI, download Advisory Board’s Generative AI Cheat Sheet.
Thomas Seay
Managing director
Posted on August 02, 2023
Updated on August 02, 2023
Don't miss out on the latest Advisory Board insights
Create your free account to access 1 resource, including the latest research and webinars.
Want access without creating an account?
You have 1 free members-only resource remaining this month.
1 free members-only resources remaining
1 free members-only resources remaining
You've reached your limit of free insights
Become a member to access all of Advisory Board's resources, events, and experts
Never miss out on the latest innovative health care content tailored to you.
Benefits include:
Unlimited access to research and resources
Member-only access to events and trainings
Expert-led consultation and facilitation
The latest content delivered to your inbox
You've reached your limit of free insights
Become a member to access all of Advisory Board's resources, events, and experts
Never miss out on the latest innovative health care content tailored to you.Benefits include:
Unlimited access to research and resources
Member-only access to events and trainings
Expert-led consultation and facilitation
The latest content delivered to your inbox
