GPT-4 based AI agents – the new expert system for detection of antimicrobial resistance mechanisms?

A study aimed to validate a customized GPT-agent to identify potential resistance mechanisms in Gram-negative bacteria. The researchers used 225 Gram-negative isolates and formed four categories based on phenotypic resistances against beta-lactam antibiotics. They customized a GPT-agent with EUCAST-guidelines, expert rules, and EUCAST-breakpoint table. The results showed that human experts showed higher concordance and shorter argumentations compared to GPT-agents. However, GPT-agents showed more unspecific flagging of ESBL and AmpC, potentially resulting in additional testing, diagnostic delays, and higher costs. The study highlights the need for validation of LLMs and datasets for benchmarking.