It was something that was studied as part of Project Plowshare, the American effort to see if nuclear weapons could be used for "peaceful" purposes like digging holes, making artificial harbors, fracking natural gas, putting out oil-well fires, and other such activities. As part of this work the two national laboratories did all sorts of studies as to how feasible it would be to use nukes to dig canals and things like that.

One of these studies — the one that I posted to Twitter awhile back, and has been seeing some attention lately thanks to the stuck ship — was from 1963, and done at the Lawrence Livermore laboratory, and is a very rough "back of the envelope" study for how many nukes it would take to dig a canal across Israel to make an alternative to the Suez canal. It does not seem to have gone much further than this rough stage, which basically involved them just measuring a route and using some simple rules to figure out how many nukes placed side-by-side would excavate it, and then coming up with a price estimate based on how much they were quoting for this kind of usage of nukes. The study itself noted that it had not really considered the political issues; it was just a basic application of some relatively simple math. It was not, as far as I know, a "serious proposal" — though if there had been interest in it, perhaps it could have been.

In the end, Project Plowshare fizzled out in the United States because the PR associated with using nukes to dig holes was still pretty negative, and it turns out that all of their candidate proposals were strongly rejected by the people living in the places that would end up being nuked. They also ran into the entire-predictable problem that digging holes with nukes creates a lot of lingering and downwind radioactivity, to a degree that it seriously reduces the utility of whatever hole you have dug.