This publication critically examines the contemporary science (CS) paradigm surrounding the bending of starlight by the Sun’s gravity, emphasizing both the theoretical and experimental challenges in validating predictions from different CS models. It questions the fundamental assumptions of the CS framework, particularly the depiction of photons as being influenced by the Sun’s gravity (Newton’s law) to bend a star’s “ray of light.” The analysis revisits historical calculations of the bending angle, starting with von Soldner’s 1804 Newtonian estimate of 0.84 arcseconds, through Einstein’s predictions of 0.83 arcseconds (1911) and 1.7 arcseconds (1916), the latter based on general relativity. The publication also explores alternative models, such as Newtonian hyperbolic trajectory calculations and Euler method approximations. In addition, experimental limitations, such as atmospheric interference, are shown to compromise the precision of measurements, including Eddington’s 1919 eclipse observations, which were widely celebrated as validation of Einstein’s theory. It further exposes a significant, overlooked inconsistency: the paradox of photon acceleration under gravity, which conflicts with Einstein’s postulate of the constancy of light speed. The irrefutable anomaly in fact falsifies the Eddington paradigm itself. The text concludes that the prevailing CS paradigm, combining Newtonian and relativistic effects to explain light bending, requires re-evaluation, as it potentially misinterprets empirical phenomena and theoretical principles.