To see why, consider that to do any useful work, data from different objects (also from different types) has to be combined. To be able to do that in the OOP framework, the encapsulation has to be unwrapped. That's why Java code is littered with getters and setters that don't do any useful work at all, they just make it too painful to get any real work done.
Again, there is a place for objects and implementation hiding, but it's at the highest levels of an architecture where different components get integrated.
This also has huge implications in a language that emphasises dynamic loading like Java. And it also flies in the face of all of the pretenses that ABI compatibility is sacrosanct and no feature that breaka it can be considered, that the design team often touts.
But I'd say that GP's complaint about inequality leaking makes no sense anyways, because what could be more unequal than different implementation, or different internal state implying different behavior down the line? The public subset isn't some arbitrary interface that could have different implementations. And even then, "equals under interface I1" would have to be considered a very special type of "equality", not the general case.