Doximity Research Reveals AI Models Struggle With Medication Doses
Doximity's research shows general-use AI models often get medication doses wrong, while its dose verification tool outperforms them. This highlights the need for accurate AI tools in healthcare.
Original reporting · healthcaredive.com
Original: “Popular AI models get medication doses wrong, some dangerously: Doximity”
Read “Popular AI models get medication doses wrong, some dangerously: Doximity” at Healthcare Dive →
How this was written. Diagnosis Forward wrote this article with a language model from Healthcare Dive’s reporting (model: @cf/meta/llama-3.3-70b-instruct-fp8-fast); it describes that reporting rather than adding to it. The original, linked above, is the full story.
Key points from Healthcare Dive’s report
- Doximity's research shows that general-use AI models often get medication doses wrong.
- Doximity's dose verification tool outperforms general-use AI models.
- The need for accurate AI tools in healthcare is increasingly important as AI plays a larger role in the sector.
Doximity, a clinical AI company, has released research indicating that popular AI models are prone to errors when it comes to medication doses. According to Healthcare Dive, this research is part of Doximity's efforts to differentiate itself as major AI companies enter the healthcare sector.
The research specifically compared the performance of Doximity's dose verification tool to that of general-use AI models. It found that Doximity's tool outperformed these models, which were shown to be rife with errors. This suggests that specialized AI tools, like Doximity's, may be better suited for tasks that require high accuracy, such as medication dosing.
The fact that general-use AI models are struggling with medication doses is significant, as it highlights the potential risks associated with relying on these models in healthcare settings. As AI continues to play a larger role in healthcare, the need for accurate and reliable AI tools becomes increasingly important. Doximity's research serves as a reminder of this need and underscores the value of specialized AI tools in improving patient care.
Questions and answers
What did Doximity's research find about general-use AI models?
Doximity's research found that general-use AI models are prone to errors when it comes to medication doses.
How does Doximity's dose verification tool perform compared to general-use AI models?
Doximity's dose verification tool outperforms general-use AI models, according to the research.
