Safety learning allows the identification of non-threatening situations, a learning process instrumental for survival and psychic health. In contrast to fear learning, in which a sensory cue (conditioned stimulus, CS) is temporally linked with a mildly aversive stimulus (US), safety learning can be studied by presenting the CS and US in an explicitly unpaired fashion during training. This should result in reduced contextual freezing when the learned CS is presented alone in a recall session within the same context. Here, we studied the effects of auditory stimuli on freezing responses in mice. Surprisingly, we found that in control experiments in which only footshocks were present during training ("US-only" controls), auditory stimuli during recall reduced ongoing contextual freezing, similar as in the regular Safety learning protocol. Upon more intensive training of mice with two training- and recall sessions, auditory stimuli induced a significant suppression of freezing both in a Safety learning group, and in US-only controls. Control experiments in the domain of fear learning, in which a new context is used during recall, did not reveal a significant reduction of contextual freezing by auditory stimuli. These experiments show that in safety learning protocols, the read-out of a presumed "safety" CS can be compromised by an intrinsic propensity of auditory stimuli to suppress ongoing contextual freezing. Thus, extensive control experiments should be performed when designing behavioral experiments for safety learning, to detect possible intrinsic effects of sensory cues on freezing behavior.