Machine Learning is about solving a task by automatically learning patterns based on examples and trying to generalize to new data. Granted, it's not necessarily the best approach to understand deeply how a phenomenon occurs.
You could use a black box model if you're more interested in predicting correctly images of handwritten digits than in understanding how the pixels relate to each other.
Of course, usually people want both accuracy and interpretability. It boils down to understanding what's more important for the problem at hand and making the compromises accordingly.
It's not that accuracy will always be sacrificed if one wants an explainable model. The point is: if interpretability is an important constraint, it could prevent improvements on accuracy.
Sometimes, the best interpretable model is as good as a black box, and that's great.
When this is not the case, the trade-off is that one should see what's more important for the actual problem. Perhaps interpretability is not a big deal.
Another solution is to try to extract interpretability from the more accurate black box model with something like SHAP.
Assuming it's céu as in the Portuguese word for sky, it's pronounced as /sɛw/ (in Brazilian Portuguese, at least). It sounds similar to "sell", only ending in /w/ rather than /l/.
This is a really nice puzzle! I didn't know William Wu wrote a paper about it. I saw it many years ago on William Wu's riddles site[1] (highly recommended). It's one of the most popular problems there. The original thread[2] in the forum has 25 pages, covering lots of approaches. There is also another thread[3] summarizing the discussion.