Background: The optimal oxygen therapy strategy for critically ill patients remains controversial. This systematic review and meta-analysis evaluated the comparative efficacy of liberal versus conservative oxygen therapy (COT) on mortality and clinical outcomes in a critically ill population. Methods: We conducted a comprehensive search of PubMed, Embase, and the Cochrane Library for randomized controlled trials (RCTs) published from database inception through December 4, 2024. Eligible studies included adult critically ill patients randomized to different oxygen therapy strategies. The primary outcome was short-term mortality (28/30-day), with secondary outcomes encompassing longer-term mortality (90 days and 180 days), healthcare-associated infections, organ dysfunctions, intensive care unit (ICU) and hospital length of stays, and mechanical ventilation parameters. Effect estimates were reported as relative risk (RR) or mean difference with a corresponding 95% confidence interval (CI). Results: Twenty-one RCTs comprising 11,901 patients were studied. COT demonstrated no significant difference in short-term mortality compared to liberal strategies (RR: 0.98; 95% CI: 0.90-1.06; P = 0.27; I2 = 19.8%). Similarly, no differences were observed in 90-day mortality (RR: 0.98; 95% CI: 0.98-1.04; P = 0.72; I2 = 0) or in 180-day mortality (RR: 0.97; 95% CI: 0.84-1.11; P = 0.89; I2 = 0). Both approaches yielded comparable outcomes regarding ICU mortality, ICU and hospital lengths of stay, new-organ dysfunctions, healthcare-associated infections, and mechanical ventilation duration metrics. The trial sequential analysis confirmed sufficient evidence to establish equivalence between the two oxygenation strategies. Conclusion: This comprehensive meta-analysis provides robust evidence that conservative and liberal oxygen therapy approaches yield equivalent mortality and clinical outcomes in critically ill patients. Clinicians should individualize oxygen therapy based on patient-specific factors rather than adopting a standardized approach.