AIR: Improved Confidence in Your AI Solutions

Modern analytic methods, including artificial intelligence (AI) and machine learning (ML) classifiers, are powerful tools that have revolutionized prediction capabilities and automation through their capacity to analyze and classify data. To produce results, most AI and ML methods depend on correlations. However, an overreliance on correlations can lead to prediction bias and reduced confidence in AI outputs.

For the past several years, the SEI has been applying and adapting novel techniques from causal discovery, which produces cause–effect graphs, and causal inference, which evaluates cause-effect relationships, to assess various classifier predictions with more nuance. These improvements have resulted in AI and ML predictions that are less biased and more suitable for guiding intervention and control of a system’s performance and better attribution of outliers and causes. The result is the AI Robustness (AIR) Tool, which allows users to gauge AI and ML classifier performance with unprecedented confidence.