Alibis play a critical role in the criminal justice system. Yet research on the process of alibi generation and evaluation is still nascent. Indeed, similar to other widely investigated psychological phenomena in the legal system - such as false confessions, historical claims of abuse, and eyewitness memory - the basic assumptions underlying alibi generation and evaluation require closer empirical scrutiny. To date, the majority of alibi research investigates the social psychological aspects of the process. We argue that applying our understanding of basic human memory is critical to a complete understanding of the alibi process. Specifically, we challenge the use of alibi inconsistency as an indication of guilt by outlining the "cascading effects" that can put innocents at risk for conviction. We discuss how normal encoding and storage processes can pose problems at retrieval, particularly for innocent suspects that can result in alibi inconsistencies over time. Those inconsistencies are typically misunderstood as intentional deception, first by law enforcement, affecting the investigation, then by prosecutors affecting prosecution decisions, and finally by juries, ultimately affecting guilt judgments. Put differently, despite the universal nature of memory inconsistencies, a single error can produce a cascading effect, rendering an innocent individual's alibi, ironically, proof of guilt. Copyright © 2017 John Wiley & Sons, Ltd.