In a case like this, I think AI might be useful in collating the data, getting the data together and making it easier to find the appropriate data. I think that's probably more where I would think it's useful.
Some AI—I haven't done that yet, maybe I should—might be useful in predicting areas that would need to be sampled or where there's a higher risk of contamination. We did a fairly thorough sampling, but it's partly random, and there are areas of the country that need to be sampled to see whether the drinking water is tainted or not.
There are definitely ways—but I'm not sure how—where AI can help better plan that sampling, because sampling is very expensive. Yes, the analysis in the lab is expensive because you need fancy instruments, but oftentimes—and it's probably even more true for my colleague—getting out there and getting the samples back to the lab is very costly.