Recently, there has been a rising interest for service dogs for epileptic people. Indeed, there are reports of dogs that are spontaneously sensitive to epileptic events of their owners. Two types of service dog are considered; dogs can be trained to perform, or spontaneously demonstrate, both roles. First, the seizure-alert dogs which seem to anticipate seizures and warn their owner accordingly. Second, the seizure response dogs which demonstrate specific behaviors during or immediately after a seizure. However, there is a lack of rigorous clinical trials to confirm these abilities and know more about the mechanisms potentially involved. The purpose of this review is to present a comprehensive overview of the validated scientific research on seizure-alert/response dogs for epileptic people. The specific goals are to a) identify the existing scientific literature on the topic, and then describe (b) the characteristics of seizure-alert/response dogs, (c) evaluate the state of evidence base, and (d) the reported outcomes of seizure-alert/response dogs. To perform a systematic review, we followed the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines of the empirical literature. Out of 28 studies published in peer-reviewed journals dealing with dogs for epileptic persons, only 5 were qualified for inclusion. Four were self-reported questionnaires, the last one being the only prospective study in the field. Reported times of alert before seizure varied widely among dogs (with a range from 10 seconds to 5 hours) but seemed to be reliable (accuracy from ≥70% to 85% according to owner reports). Alerting behaviors were generally described as attention getting. The alert applied to complex partial and generalized tonic-clonic seizures as well as atonic and absence seizure. Pet dogs as dogs recruited as seizure-alert dogs generally varied in size and breed. Training methods differed widely between service animal programs, partially relying on hypothesized cues used by dogs (e.g : variations in behavior, scent, heart rate, etc.). However, none of the studies included a comparison condition, and only one looked at the treatment condition, using a pre-post design and a follow-up assessment. There was a low level of methodological rigor in most studies (e.g. regarding medical diagnosis, self-reporting bias, choice of study design, etc.), indicating the preliminary nature of this area of investigation. Furthermore, most studies indicated an increase in quality of life when living with a dog demonstrating seizure-related behavior. In addition, owner reports suggest that the presence of a dog with seizure-related behavior had an impact on epilepsy, reducing the seizure frequency of epileptic persons. In conclusion, scientific data are still too scarce and preliminary to reach any definitive conclusion as to the success of dogs in alerting that a seizure will come, the cues on which this ability may be based, the type of dog that could be most successful and the type of training that should be done. While these preliminary data suggest that this is a promising topic, further research is needed. This argues for appropriate and controlled prospective studies on dogs prediction abilities, for example with video-EEG designs.