The IBM watsonx.governance advantage

Our scope of use is wide but deliberate. At first, I introduced it through a pilot in our scrap metal forecasting project, which predicts how much reusable steel can be reintroduced into prod each quarter. There were other multiple layers involved before running the outputs through IBM watsonx.governance to enforce governance on the data lineage. Since then, we've expanded it into supply chain risk modelling.

Policy automation around ai models. We run these on our rolling mills. I've set up automated checks that ensure every retained model logs its inputs and validation results

I could go into the technical bits but in all honesty, I'm most thankful to the fact that I can now walk into compliance and audit meetings with confidence rather than a binder full of spreadsheets.
I also do very little of firefighting, since i don;t have to scramble through disconnected datasets to justify why our ai flagged a risk