BACKGROUND: Serious games (SGs) are nonpharmacological interventions that are widely applied among older adults. To date, no evidence has been published regarding the effect of digital SGs on cognitive ability, daily behavioral capacity, or depression in older adults with Alzheimer's disease (AD) and mild cognitive impairment (MCI). OBJECTIVE: This study aimed to assess the effect of SGs on older adults with AD and MCI by summarizing and pooling the results of previous studies. METHODS: This meta-analysis examined the effectiveness of digital SGs in improving cognitive ability, enhancing daily behavioral capacity, and alleviating depression in older adults with AD and MCI. We searched the following databases up to December 31, 2023, to identify relevant high-quality randomized controlled trials (RCTs): PubMed, Embase, Web of Science, Scopus, and Cochrane Library. Stata 15.1 and Review Manager 5.3 were used to screen the 14 studies, extract data, code the data, and perform meta-analysis. Mean differences and standardized mean differences (SMDs) with 95% CIs were used to calculate continuous variables. The Cochrane risk-of-bias assessment tool was used to evaluate the risk of bias. Eligibility criteria were developed in accordance with the Population, Intervention, Comparison, Outcomes, and Study Design framework: (1) population (older adults with AD and MCI), (2) intervention (digital SG intervention), (3) comparison (digital SG intervention vs routine health care), (4) outcomes (cognitive ability, daily behavioral capacity, and depression), and (5) study or research design (RCT). Sensitivity analysis was performed, and a funnel plot was constructed. RESULTS: From January 2017 to December 2023, we enrolled 714 individuals across 14 RCTs, with 374 (52.4%) in the severe game group using digital SGs and 340 (47.6%) in the control group using traditional methods. The results of our meta-analysis indicated that using digital SGs in older adults with AD and MCI is more effective than traditional training methods in several key areas. Specifically, digital SG therapy significantly increased cognitive ability, as found in the Mini-Mental State Examination (SMD 2.11, 95% CI 1.42-2.80; P<.001) and the Montreal Cognitive Assessment (SMD 2.75, 95% CI 1.98-3.51; P<.001), significantly increased daily behavioral capacity (SMD 0.53, 95% CI 0.06-0.99; P=.03), and significantly reduced depression (SMD -2.08, 95% CI -2.94 to -1.22; P<.001) in older adults with AD and MCI. No publication bias was detected based on the results of Begg and Egger tests. CONCLUSIONS: Digital SGs offer a viable and effective nonpharmacological approach for older adults with AD and MCI, yielding better results compared to traditional formats. However, caution is warranted in interpreting these findings due to limited RCTs, small sample sizes, and low-quality meta-analyzed evidence. TRIAL REGISTRATION: PROSPERO International Prospective Register of Systematic Reviews: CRDCRD42023486090; https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID=486090.