STUDY QUESTION: What is the true incidence of chromosomal mosaicism in embryos analyzed by preimplantation genetic testing (PGT) SUMMARY ANSWER: The true incidence of chromosomal mosaicism is much lower than we usually surmise. WHAT IS KNOWN ALREADY: In recent years, contemporary methods for chromosome analysis, along with the biopsy of more than one cell, have given rise to an increased rate of chromosomal mosaicism detection after preimplantation genetic testing for aneuploidy. However, the exorbitant incidence of mosaicism represents a dilemma and imposes restrictions on the application of PGT treatment. Concern has been raised about the possibility that the incidence of chromosomal mosaicism is overestimated and quite a few of the results are false-positive errors. However, studies verifying the diagnosis of chromosomal mosaicism and assessing the true incidence of chromosomal mosaicism are limited. STUDY DESIGN, SIZE, DURATION: A total of 1719 blastocysts from 380 patients who underwent PGT treatment were retrospectively analyzed to evaluate the typical incidence of mosaicism. Then 101 embryos donated by 70 couples were re-biopsied and dissected into three portions if available: trophectoderm (TE), inner cell mass (ICM), and the remaining portions. All the portions were tested using next-generation sequencing (NGS), and the results were compared to the original diagnosis. PARTICIPANTS/MATERIALS, SETTING, METHODS: The setting for this study was a university-affiliated center with an in-house PGT laboratory. All samples were amplified with multiple annealing and looping-based amplification cycles (MALBACs) and the NGS was carried out on a Life Technologies Ion Proton platform. MAIN RESULTS AND THE ROLE OF CHANCE: A clinical TE biopsy revealed an incidence of 11.9% for diploid-aneuploid mosaicism (DAM), 17.3% for aneuploid mosaicism (AM) and 29.1% in total. After rebiopsy, 94.1% whole-chromosome aneuploidies and 82.8% segmental-chromosome aneuploidies were confirmed in the embryos. As for the mosaic errors, only 32 (31.7%) out of 101 embryos presented with uniform chromosomal aberrations in agreement with the original biopsy results, 15 (14.8%) embryos presented with de novo chromosomal aberrations, and 54 (53.5%) embryos showed a euploid profile in all portions. Among the 32 uniform embryos, the true mosaicism was confirmed in only 4 cases, where a reciprocal chromosomal aberration was identified; 14 embryos presented with identical mosaicism, providing the moderate evidence for true mosaicism; and 14 embryos displayed uniform full aneuploidies in all portions of embryo, revealing a high-grade mosaicism or a false-negative diagnosis. Logistical regression analysis revealed that the concordance rate with ICM was associated with the type and level of mosaicism. The concordance rate of segmental-chromosome mosaicism was significantly lower than whole-chromosome mosaicism (adjusted Odds Ratio (aOR): 5.137 (1.061, 24.876), P = 0.042) and compared to DAM, the concordance rate of AM was significantly higher (aOR: 6.546 (1.354, 31.655), P = 0.019). The concordance rate also increased with increasing levels of mosaicism (P<0.001). LIMITATIONS, REASONS FOR CAUTION: This study was limited by a small sample size and the use of a single whole-genome amplification (WGA) method and NGS platform. These findings are only applicable to samples subjected to MALBAC amplification and Ion Proton platform, and studies involving larger sample sizes and multiple WGA methods and NGS platforms are required to prove our findings. WIDER IMPLICATIONS OF THE FINDINGS: TE biopsy is reliable to detect whole-chromosome aneuploidies, but the ability to diagnose mosaicism is doubtful. More attention should be paid to false-positive and false-negative errors in NGS-based PGT, especially for laboratories using less stringent criteria for mosaicism classification (i.e. 20-80%), which might be subject to a much higher false-positive mosaicism rate in the practice.