Consider this hierarchy: Natural/Artificial(whatever these terms mean) -> Genetic Modification(General field) -> Plant Transformation(Specific method) -> Using

Consider this hierarchy: Natural/Artificial(whatever these terms mean) -> Genetic Modification(General field) -> Plant Transformation(Specific method) -> Using Agrobacterium tumefaciens(Specific technique) -> The application in practice. Even if one proves that something applies to a step/layer, one cannot generalize downwards (towards the general); only upwards (towards the specific). That's why I find the anti-GMO craze weird: Genetic modification in itself isn't harmful, and also happens in nature all the time (as if this makes a difference). It's specific outcomes of certain modifications, or side-effects of specific applied techniques (as noted in the article you linked to). So, whether a genetic modification is performed by "chance" (as in by nature), or deliberately (as in by humans) is irrelevant in and of itself. Weakness or risk is inherent not in the performing agent (or lack thereof), but in the the technique or its application – which is clear from the article.