I think this is great and as a practitioner, something I've incorporated and demand in all of the contracts I work on is a clear and explicit validation and unc

I think this is great and as a practitioner, something I've incorporated and demand in all of the contracts I work on is a clear and explicit validation and uncertainty analysis. It was always an issue when I worked at big corp that there was pressure to bury the lede when it came to uncertainty/ propagation of error. I considered it to be misleading and since then I make sure that any work that I'm a part of has in explicit language exactly how we'll analyse and approach bias and uncertainty as a part of the original work. I also try and make sure that assessments of uncertainty are 'part of the product' so there is no getting around things and every prediction also comes with an 'I told you so' layer. I'm an absolute proponent of the kinds of products that get labeled as AI. I also think that useful predictions can be made in the face of uncertainty, because lets be real, human assessments, predictions and decisions also come with a degree of uncertainty (we usually just fail to rigorously quantify them).