The landmark ARDS Network trial compared a lung-protective ventilation strategy using lower tidal volumes with a traditional larger-tidal-volume approach in patients with acute lung injury and acute respiratory distress syndrome. The lower-tidal-volume strategy reduced ventilator-induced lung stress and was associated with lower mortality and more ventilator-free days. The study established that the ventilator itself can worsen an already injured lung when excessive volumes are delivered. Its influence extends well beyond ARDS: modern critical care increasingly applies the principle of avoiding unnecessary mechanical stress, with tidal volume related to predicted rather than actual body weight and airway pressures monitored closely. The trial also showed that accepting a degree of hypercapnia may be preferable to exposing the lung to injurious volumes, provided there is no contraindication. For practicing clinicians, the key message is simple but powerful: mechanical ventilation is supportive therapy, but it can also cause harm. The safest strategy is therefore to ventilate the smallest functional lung volume that achieves acceptable gas exchange while monitoring plateau pressure and other markers of respiratory mechanics. This paper remains the cornerstone for interpreting subsequent work on driving pressure, PEEP, prone positioning, neuromuscular blockade and extracorporeal support.