Because Annila says "when light passes by a local energy-dense area, such as a star, the speed of light will change and its direction of propagation will change". Einstein said this: "I arrived at the result that the velocity of light is not to be regarded as independent of the gravitational potential. Thus the principle of the constancy of the velocity of light is incompatible with the equivalence hypothesis". To back that up, Einstein also said this: "In the second place our result shows that, according to the general theory of relativity, the law of the constancy of the velocity of light in vacuo, which constitutes one of the two fundamental assumptions in the special theory of relativity and to which we have already frequently referred, cannot claim any unlimited validity. A curvature of rays of light can only take place when the velocity of propagation of light varies with position".
Uh... yes? You place too much importance on a very vague similarity.
When you press them on what the guy is saying and some other guy shows an important link to Einstein's GR which he allegedly contradicts, what you then get is essentially I don't care what he says, followed by Latin, followed by burn him.
The important link is, what, that light bends because the gravitational potential acts somewhat analogously to a variable index of refraction? Take a look at
this wikipedia page. All those theories of gravity for which γ is not -1? The exact same thing happens in all of them. It is not some deep link between Einstein and Annila; it's a basic feature of theories of gravity which couple to light. For Φ = -GM/rc², just as for Annila's massive body case he discusses at the beginning,
[latex]$ds^2 = -(1+2\Phi+2\beta\Phi^2+\ldots)(c\,dt)^2 + (1-2\gamma\Phi+\ldots)dS^2$[/latex]
Therefore for light, ds = 0 gives:
[latex]$n^{-2} = (dS^2/dt^2)/c^2 = 1 + 2(\gamma+1)\Phi + 2(\beta+2\gamma(\gamma+1))\Phi^2 + \ldots$[/latex]
So Annila's theory forces γ = -1/2 and, which is ruled out in several ways: angular deflection of light discussed previously, Shapiro delay (e.g., Cassini), and gyroscopic precession (e.g., Gravity Probe B), and probably some other ways I'm ignorant of. GTR requires this to be 1.
In other words:
even if one were to take Annila's handwaving of why we don't see a corresponding increase of deflection for the Sun as a given, it still
just doesn't work, because the angular deflection if not the only thing this change affects. There's a potential loophole as to whether Annila's theory predicts β = 1/2, but if it does, the number of types of observations that contradict it becomes 'just about everything'.