If every quantum outcome happens in some branch, then somewhere a version of you survives everything. Subjectively, you might only ever experience the branches where you live.
Documented
The thought experiment, stated and disclaimed
Tegmark's 1997 proposal: if many-worlds is true, an experimenter facing a quantum-triggered lethal device with 50-50 odds per run would, after many runs, find themselves alive with certainty in their own experience, since dead branches contain no experience. Everett's daughter noted that her father believed in a kind of quantum immortality and was untroubled by death. Tegmark himself listed three reasons the trick fails in practice: real death is rarely quantum-random and is usually gradual, not instantaneous branching.
Assessment: The logic is documented as a conditional: given many-worlds and instantaneous quantum-dependent death, subjective survival follows. Many-worlds remains an interpretation with no confirming experiment, and the failure modes are documented by the author himself.
Stories
The internet's favorite afterlife
Online, the idea shed its caveats and became folklore: you have already survived things that should have killed you, near misses as branch evidence, the dead just wake up in the next timeline. It functions as a secular afterlife with equations as scripture, and it comforts exactly like one.
Assessment: The folk version is documented belief with no evidence behind it. Surviving a near miss proves the near miss; the billions of counterfactual deaths are invisible by construction, which is not the same as real.
Interesting claims
Identity as a river that never hits the sea
The interesting claim generalizes: even in ordinary aging, death by degrees means there is always some branch where the pattern continues, so the stream of subjectivity never experiences its own end; everyone dies in everyone else's story and no one dies in their own. It is the measurement problem's creepiest corollary.
Assessment: This is metaphysics stacked on an unconfirmed interpretation, untestable by design since only the subject could verify it, at the price of being unable to report. As consolation or horror it is genuine; as physics it is a shadow cast by an equation's meaning, not a finding.
Sources and further reading
Tegmark, The Interpretation of Quantum Mechanics: Many Worlds or Many Words?, Fortschritte der Physik (1998), the thought experiment with its caveats
Everett, Relative State Formulation (1957), the premise
Byrne, The Many Worlds of Hugh Everett III (2010), including the immortality belief
Deutsch, The Fabric of Reality (1997), the strongest many-worlds advocacy
Tegmark, Our Mathematical Universe (2014), the worldview at full strength
Worth asking
The argument grants you survival in some branch while everyone you love watches you die in theirs. Is that immortality, or the loneliest idea ever formalized?