Abstract Improved predator detection is often stated to be one of the principal benefits of social foraging. However, actual field evidence supporting this assumption is scarce. This may be the result of the fact that most observations are conducted on social animals acting in the absence of an acute predation threat, yet the benefits of grouping come to the fore in that brief moment when an individual's life is at risk. As predation attempts are typically rare in nature, experimental manipulations are necessary to further explore the costs and benefits of social foraging. This study utilizes simple predator simulations (by humans) to experimentally test the predator-detecting abilities and escape strategies of groups of free-living emus Dromaius novaehollandiae. Emus in larger groups spent less time in vigilance and more time foraging. Nonetheless, the combined vigilance of group members ensured that emus detected the ‘simulated predator’ sooner as group size increased. After detecting the ‘predator’, larger groups waited longer until opting to flee, and then spent less time and energy doing so. Thus, the results of this study provide experimental evidence that emus benefit from grouping in terms of both the ‘many-eyes effect’ and the ‘dilution effect’.