IBM Cracks AI's 'Black Box'

IBM has introduced technology intended to provide businesses new transparency into AI. The software service, which automatically detects bias and explains how AI makes decisions—as the decisions are being made —runs on the IBM Cloud, and is designed to help organizations manage AI systems from a wide variety of industry players. In addition, IBM Research will release into the open source community an AI bias detection and mitigation toolkit, bringing forward tools and education to encourage global collaboration around addressing bias in AI.

IBM's new Trust and Transparency capabilities on the IBM Cloud work with models built from a wide variety of machine learning frameworks and AI-build environments such as Watson, Tensorflow, SparkML, AWS SageMaker, and AzureML. This means organizations can take advantage of these new controls for most of the popular AI frameworks used by enterprises.

The software service can also be programmed to monitor the unique decision factors of any business workflow, enabling it to be customized to the specific organizational use. The automated software service explains decision-making and detects bias in AI models at runtime – as decisions are being made—capturing potentially unfair outcomes as they occur. Importantly, it also automatically recommends data to add to the model to help mitigate any bias it has detected.

Explanations are provided in easy to understand terms, showing which factors weighted the decision in one direction vs. another, the confidence in the recommendation, and the factors behind that confidence. Also, the records of the model's accuracy, performance and fairness, and the lineage of the AI systems, are easily traced and recalled for customer service, regulatory, or compliance reasons—such as GDPR compliance.

All of these capabilities are accessed through visual dashboards, providing business users an ability to understand, explain and manage AI-led decisions. IBM said it is also making available new consulting services to help companies design business processes and human-AI interfaces to further minimize the impact of bias in decision making. 

IBM Research is making available to the open source community the AI Fairness 360 toolkit—a library of novel algorithms, code, and tutorials that will give academics, researchers, and data scientists tools and knowledge to integrate bias detection as they build and deploy machine learning models.

For more information, go to