Assessment of the efficacy of ChatGPT responses to bacterial species-specific questions in microbiology. | ScholarDeck

Paper

Assessment of the efficacy of ChatGPT responses to bacterial species-specific questions in microbiology.

New research · AI in Medicine

Access microbiology · 3w

AI / informaticsAccess microbiology· 2026

Assessment of the efficacy of ChatGPT responses to bacterial species-specific questions in microbiology.

Withanage Dona Manushi Dinasha, Nissanka Mudiyanselage Tanuri Ayanga Nissanka, Chamudhi Prabashi Wickramasinghe … Hiripitiyage Gayan Danushka Gunatilake

Read paper

Large Language Models in CareAI / informaticsQ3IF 1.4

ChatGPT 4.1 mini gave accurate answers to bacterial species questions more often than ChatGPT 3.5

Assessment of the efficacy of ChatGPT responses to bacterial species-specific questions in microbiology.

Withanage Dona Manushi Dinasha … Hiripitiyage Gayan Danushka Gunatilake

Access microbiology · 2026

Background

Background.

Purpose

Objectives.

Results

4.1 mini answered more bacterial species questions accurately than 3.5

ChatGPT 3.5 accurate

40.4%

ChatGPT 4.1 mini accur

56.3%

More results

However, assessing the accuracy of ChatGPT's responses is essential due to the potential for 'hallucinations' in large language models (large language models).

A clinical microbiologist finalized a list of 15 bacterial species, each with 18 specific questions of both local and international relevance.

More results

These questions were then prompted to ChatGPT 3.5 and 4.1 mini models, simulating real user interactions.

Notably, ChatGPT 4.1 mini gave clearer and more reliable answers than ChatGPT 3.5.

“

Conclusion

The study emphasizes the influence of question formulation on response accuracy, recommending further research to explore more advanced large language models like ChatGPT-4o and ChatGPT-o3 models.

Read paper