Ensuring Trust in AI-Driven Chromatography

Dave Abroamowitz of Thermo Fisher Scientific concludes his five-part video series with thoughts on how laboratories can address potential biases or errors introduced by artificial intelligence (AI) models, especially when these analyses inform high-stakes decisions such as clinical research or regulatory submissions.

ABRAMOWITZ: You have to have a human in the loop who has to do that gut check, that sanity check on a lot of these things, and ensure that the models that the people are building are actually tested properly. Testing is one of the hardest things to actually get… so creating that testing data is one of the ways that we can ensure we can remove errors and bias in the models we are building.