My practical approach trumps all unprovable scientific theories that cannot be demonstrated. Seeing is believing.
“No lesson is so deeply inculcated by the experience of life as that you should never trust experts.” Lord Salisbury
"Question everything." TV Science channel
Of course we should question everything, and not blindly trust experts. In a way thats the basis for the scientific method, that we repeat experiments and studies over and over again because we dont take someones word for it. That said, what appears to be true can be very far from the truth. For example, take the heliocentric and geocentric models of the solar system. One of the reasons the geocentric model was dominant in the passed was because the theories based on it gave better predictions of planetary movements. Then someone, think it was Kepler, realized that planets didnt have circular orbits, but elliptical. Thats why the geocentric model gave better results earlier, not because it was true, but because the heliocentric people viewed the orbits as circular and, as such, their calculations didnt match the actual movements. The geocentric model managed that much better by putting planets in orbits that can be described as circles in circles in circles (or something like that). If I remember correctly, Keplers idea of elliptic orbits proved to not only give better results, but it was also vastly simpler. Newton later derived Keplers laws of planetary motion from his laws. Bottom line is, no, we shouldnt blindly accept truths from experts. Kepler and Newton didnt. But dont assume there isnt something to what the experts say either. Newtonian mechanics has shown to be quite accurate in many situations. There are places where its not practical, mainly at the very small scales or where relativistic effects become noticable (very high speed and strong gravitational fields).
Also, if I remember my class in machanics correctly, Newtons first law is actually just a consequence of Newtons second law, F = ma. Essentially it means that it takes a force for velocity to change, which to my knowledge is backed up by observations.