To be detected, as signal must be significantly stronger than noise that would interfere with detection. Detecting photons is not the issue, detecting SIGNALS is.

All visible stars produce microwaves as well as visible light. If you are looking at a stellar system, you will necessarily have the STAR ITSELF in your sights. This will blot out any small signal that you might receive from an unbeamed broadcast. Only a narrow beam can put more photons of any chosen band per square degree than the star will always be doing.

I therefore conclude that the chance of detection of an unbeamed (spherical) signal is zero, UNLESS some way is found to occult the central star of the system. There have been proposals to do this with various optical tricks, but so far it has not been done. Only such an aparatus could, in principle, detect signals that were not beamed in our direction.

There is always the random chance that we will HAPPEN to be in the sweep of a beam, but that is an absurdly long shot.