A novel approach on the problem of hot cracks in Cu-containing steels is proposed to prevent the hot shortness. Cu-containing steels have been hot tensile tested to failure in argon or vacuum at different cooling rates and temperatures, followed by Scanning Electron Microscopy. EDX analysis both in fractures and cross section show that inclusions are very important in causing the Cu concentration. Although CuS precipitates slightly deteriorate the hot ductility between the Ae 3 and Ar 3 temperatures, the Cu concentration around inclusions can highly contribute to hot shortness by the formation of Cu-containing liquid phase. Laboratory results combine with thermodynamic calculations show that a Cu-containing liquid associated with inclusions at %1200 8C is clearly the cause of the hot shortness. A distinction of hot ductility and hot shortness is proposed, concerning to their very different cracking mechanisms. The interpretation of the hot tensile test for each testing condition and temperature range is discussed. It is concluded that the oxidation is not the principal cause of the increase of Cu concentration, so its importance is relative in the laboratory assessment of hot cracking. Consequently, inclusions cannot be ignored in assessing and preventing the hot shortness, and commercial implications are drawn from the results.