Microsoft Research today open-sourced a software toolkit aimed at solving AI’s “black box” problem. InterpretML is made to help developers experiment with ways to introduce explanations of the output of AI systems. InterpretML is currently in alpha and available on GitHub.

The ability to interpret predictions made by AI systems has become of increasing concern as AI is applied more frequently in sectors like law and health care.

“If doctors, judges and other decision makers trust the models that underpin intelligent systems, they can make better decisions. More broadly, with fuller understanding of models, end users might more readily accept the products and solutions powered by AI, while growing regulator demands might be more easily satisfied,” Microsoft data scientists and engineers said in a blog post today.

Microsoft researchers also believe explainability can help developers make decisions about the best approach to take to train an AI model or gauge consistency between methods.

The move comes as Microsoft wraps up its annual Build developer conference in Seattle, where the company introduced the Fluid framework for document creation on the web, additions to the Azure Kubernetes Service, and the first pieces of an AI and robotics toolkit.

Ahead of Build, last week Microsoft announced its Open Neural Network Exchange (ONNX) now supports Nvidia’s TensorRT and Intel’s nGraph for high-speed inference on Nvidia and Intel hardware.

Microsoft is a member of the Partnership on AI, a group of more than 80 tech companies and nonprofits that recently declared that algorithms are not suitable for automating the pretrial bail process in part due to the lack of a clear explanation of how plaintiffs are labeled as high risk or low risk.

Azure cloud and AI head Scott Guthrie asserted last week that Microsoft’s willingness to participate in open source projects goes further than Amazon or Google.

“I think you’re seeing a Microsoft that’s both deeply embracing openness, both as consumers, but also as contributors, and I think that’s unique. If you look at, say, AWS’ contributions to open source, there’s not a lot. There’s a lot of consumption, but there’s not a lot of contribution back, and I think even if you were to look at Google relative to the amount of contributions we’ve made on Azure, I think people are often pleasantly surprised when they add it up.”

VentureBeat has asked for more details on how the tool is used internally at Microsoft. This story will be updated if we hear back.