Nature 474, 356–358 (2011); doi:10.1038/nature10103

Subsequent analysis with updated methodology by us and others1,2 has not confirmed the detection of the population described in this Letter. We suspect, as described in later work by our group3, that the original signal was likely to be an artefact arising from the cut we chose for the background subtraction for the stacked archival data (our technique and adopted methodology is described in detail in the Supplementary Information of the original Letter). The technique of stacking and subtracting the background for Chandra data has since been much refined, and therefore our original Letter can now at best be interpreted as providing guiding upper limits. The conclusive detection of an observational signal from an early population of black holes awaits new data and possibly new instruments, for instance those onboard the upcoming James Webb Space Telescope (JWST) mission and Athena and other planned X-ray observatories. The original Letter has not been corrected.