P8 went pretty well. Blended wing isn’t sci-fi, B2 and plenty of ucav are full flying wing.
Did it really?
According to this fairly recent GAO report[1], the Poseidon's mission capable rates have missed their minimum 80% target since 2015, trending downwards (Table 1); only the first 2 out of 7 years were satisfactory (Figure 7), and yet those years' metrics were based on Increment 1 low-rate initial production lot that were "not effective for the intelligence, surveillance and reconnaissance mission" and "not effective for wide area anti-submarine search" that they were originally intended for[2].
Even in early years when maintenance staff was stacked balls deep at well over 100% (Figure 5), persistent supply issues plagued the platform, trending upwards (Table 5). To be sure, this outcome was entirely predictable; e.g. picking on Northrop Grumman's Electronic Support Measures system[3] that the Bloomberg article calls out as just one example: if you knew how the sausage was made, then this 2006 GAO report[4] which highlights certain risks "of particular concern" surrounding ALQ-218 development is all of sudden prescient in that it took so long to get the underlying system to acceptable maturity on the Growler that it became a very real commercial obsolescence sustainment risk by the time the Poseidon's full-rate production decision was made!
[1] https://www.gao.gov/products/gao-22-104533
[3] https://news.northropgrumman.com/news/releases/northrop-grum...