Well no, from the context of my remarks, it should be reasonably clear that the prevalent modern view of QM is that "observation", in the sense of perception by a conscious observer, does not affect reality. What is often called "observation" in QM, e.g. when speaking of the collapse of the wave function and so forth, is really interaction with a measuring system of some kind. This is independent of whether or not a conscious observer is aware of the measurement or not. It is true, as you say, that all observation involves interaction. However the converse does not follow. Many interactions go unobserved, after all.