Since we are talking about gauginos, I will report another very recent paper which I believe adds some plausibility to Alejandro's ideas, without telling us exactly how to realize them.
http://arxiv.org/abs/1105.1510" that "The reader will recognize that our magnetic spectrum resembles a supersymmetric one, however, the magnetic theory is not supersymmetric since we do not invoke supersymmetric relations among its spectrum and couplings."
In this http://arxiv.org/abs/1105.1510" , they study various fixed points of this nonsupersymmetric dual, and find one where the gaugino coupling evolves like the gauge coupling; so they call it emergent supersymmetry. It must be the adjoint quark, "lambda", which is the emergent gaugino in their model (the paper does not spell this out). It's not the field that is emergent, it's the behavior of the coupling; it's because the coupling of the "lambda" field matches the coupling of the gauge field that it acts like a gaugino field.
I have never had much of an opinion about the gauginos in Alejandro's model, because we don't see any such particles. What has impressed me is the matching among known objects. I was not (and am not) convinced that the theoretical realization of the correspondence will necessarily involve gauginos. In the gauge-gauge duality here, there is an adjoint quark on both sides of the duality, so it's not as if we have an emergent adjoint quark, constructed from other variables. Apart from the change in the gauge group, what happens in the duality is that the fundamental quarks in the "electric" theory are replaced by new fundamental quarks in the "magnetic" theory, and one or two extra fields appear. In other words, for this emergent supersymmetry via Seiberg duality to occur, one still has to have matter in the adjoint representation on both sides of the duality, and no such matter is known in the real world.
Then again, I have tended to think of Alejandro's correspondence as a way to realize supersymmetry which is wholly disjoint from the usual conception, in which we keep looking for massive superpartners at ever higher energies. Instead, I took the point to be that some form of supersymmetry might already exist right in front of us, connecting Standard Model elementary fields with certain QCD composites. I suppose it's possible to take a hybrid approach, and say that the Standard Model already contains supersymmetry, but that some of the usual bestiary of massive superparticles (in this case, gauginos) do remain to be discovered, as new fundamental degrees of freedom at higher energies.
So that we don't stray entirely from the original topic of this thread (which is an important topic), let me point out the reference, on page 4 of 1105.1510, to "minimal walking technicolor models", which are said to have similar properties and to be of phenomenological interest. It's asserted in http://arxiv.org/abs/hep-ph/0607191" that 'Higgsless models may be viewed as dual to models of dynamical symmetry breaking akin to “walking technicolor”' - they mean Higgsless models in which unitarity of W-W scattering is preserved by the existence of W' bosons, which can arise as KK states in an extra dimension, or in some related way in a "deconstruction" model where the existence of an extra dimension is approximated by having a discrete number of copies of the d=4 theory (the copies are like slices through a discretized fifth dimension... that is, in the continuum limit the fields exist throughout five dimensions, but when the fifth dimension is discretized into a stack of surfaces, you get those fields existing on each discrete surface, so it's as if you have a set of copies of the four-dimensional version of the fields).
Maybe you could even connect deconstruction to this newly-discovered "eta-function regularization"... ;-)