How the system actually works
Both wireless (radio-signal, circular boundary) and traditional buried-wire invisible fences use the same basic mechanism: the dog wears a receiver collar, and crossing the set boundary triggers a warning beep followed by a static correction if they continue. The correction is a form of aversive stimulus, mild for most dogs, but genuinely unpleasant enough to be effective, which is the whole point of the system.
Because it's not a physical barrier, containment relies entirely on the dog learning to respect an invisible line, which only happens through deliberate training, not just wearing the collar.
Why the training period isn't optional
Manufacturers include boundary flags for a reason: dogs need to be taught, over a period of typically two to four weeks, to associate the flags (and eventually the beep alone) with the boundary, walked to the edge, warned, redirected back, and rewarded for staying inside. Skipping this step and just letting a dog outside with the collar on is a common cause of both containment failure and unnecessary correction, since the dog has no way to predict or understand the boundary yet.
Done properly, most dogs learn to respond to the warning tone alone and rarely receive an actual correction once trained, the tone becomes the meaningful signal, similar to how a marker word works in positive training, just built on an aversive rather than a reward.
What it doesn't do
An invisible fence is one-directional: it can (imperfectly) keep your dog in, but it does nothing to keep other dogs, wildlife, or people out of your yard. A loose dog, an off-lead visitor's dog, or a person walking onto your property can all still reach your dog, who has no physical barrier for protection, a real safety consideration, especially for reactive or fearful dogs.
There's also a known failure mode where a highly motivated dog (chasing wildlife, reacting to another dog) will push through the correction to get out, and then be reluctant to cross back in the other direction because the correction fires both ways, leaving them stuck outside the boundary and distressed.
The welfare and humane-use question
Because containment depends on a dog reliably associating the warning tone with an unpleasant consequence, poorly trained or inconsistent use can create anxiety around the yard itself, rather than clean boundary awareness. Dogs that are anxious, noise-sensitive, or already reactive are generally worse candidates for this system, since an aversive correction can compound existing stress rather than simply teach a boundary.
If you're considering one, a proper training period with flags, consistent short sessions, and close supervision throughout is the difference between a system that works humanely and one that just adds stress without reliable containment. A physical fence, where feasible, remains the option with the fewest welfare trade-offs and the only one that protects against threats coming in, not just your dog going out.