I don't completely agree Willy. If you exclude water from the bottom of an upturned cup and place it on the bottom, it will stay there.

This assumption only holds true while the inner vessel is in contact with the outer vessel
It is not a true exception to  Archimedes principle.
But rather that multiple vessels must be taken into consideration

This downward force then increases linearly with depth, which clearly contrasts to the force predicted by AP, but agrees to experimental results [2, 15, 17]. In fact, the increase of this force with depth has been subject of deeper discussions in recent works [2, 15, 16, 33], in which it is suggested that the meaning of the word ‘immersed’ should be ‘fully surrounded by a liquid’ instead of ‘in contact to a liquid’, which would make the ‘bottom’ case, as well as all other ‘contact cases’, out of scope of the Archimedes original propositions, as well as AP modern statement [2, 15, 33]. Note, however, that this redefinition is deficient because it excludes some common cases of buoyancy such as, for instance, that of a solid (e.g., a piece of cork) floating in a denser liquid (e.g., water). In this simple example, the body is not fully surrounded by a liquid and yet AP works! More recently, other authors have argued that the definition of BF itself should be changed to “an upward force with a magnitude equal to the weight of the displaced fluid” [16]. However, I have noted that this would make AP a definition for the BF and then, logically, AP would not admit any exceptions at all. In face of the downward BF experiments already mentioned, it is clear that this is not a good choice of definition. Therefore, I would like to propose the abandon of such redefinitions, as they are unnecessary once we admit some exceptions to the AP, which is the natural way to treat the exceptional cases not realized by Archimedes in his original work.

Could you elaborate a little?

In all my years of searching, I have never seen a successful OU device, mechanical, magnetic or electrical.