The optimal threshold of controlled attenuation parameter (CAP) for the detection of hepatic steatosis using both M and XL probe is unknown in nonalcoholic fatty liver disease (NAFLD). Magnetic resonance imaging proton density fat fraction (MRI‐PDFF) is an accurate and precise method of detecting the presence of hepatic steatosis that is superior to CAP. Thus, the aim of this study was to evaluate the diagnostic accuracy and optimal threshold of CAP for the detection of hepatic steatosis as defined by MRI‐PDFF ≥ 5%. This prospective cross‐sectional study included 119 adults (59% women) with and without NAFLD who underwent MRI‐PDFF and CAP using either M or XL probe when indicated within a 6‐month period at the NAFLD Research Center, University of California, San Diego. The mean ( ± standard deviation) age and body mass index were 52.4 (±15.2) years and 29.9 (±5.5) kg/m², respectively. The prevalence of NAFLD (MRI‐PDFF ≥ 5%) and MRI‐PDFF ≥ 10% was 70.6% and 47.1%, respectively. The area under the receiver operating characteristic (AUROC) of CAP for the detection of MRI‐PDFF ≥ 5% was 0.80 (95% confidence interval [CI], 0.70‐0.90) at the cut‐point of 288 dB/m and of MRI‐PDFF ≥ 10% was 0.87 (95% CI, 0.80‐0.94) at the cut‐point of 306 dB/m. When stratified by the interquartile range (IQR) of CAP, we observed that an IQR below the median (30 dB/m) had a robust AUROC compared with an IQR above the median (0.92 [95% CI, 0.85‐1.00] versus 0.70 [95% CI, 0.56‐0.85]; P = 0.0117), and these differences were statistically and clinically significant. Conclusion: The cut‐point of CAP for presence of hepatic steatosis (MRI‐PDFF ≥ 5%) was 288 dB/m. The diagnostic accuracy of CAP for the detection of hepatic steatosis is more reliable when the IQR of CAP is <30 dB/m. These data have implications for the clinical use of CAP in the assessment of NAFLD. (Hepatology 2018;67:1348‐1359)