The article questions the reliability of the estimate that 10 million people will die annually from antimicrobial resistance (AMR) by 2050, as presented in the UK government's "Antimicrobial Resistance: Tackling a crisis for the health and wealth of nations" report. The authors argue that the estimates are based on flawed assumptions and lack transparency, making them unreliable. They highlight several issues, including the use of data from a non-representative surveillance network (EARS-Net), which only includes data from tertiary hospitals, and the extrapolation of these data to the entire population. Additionally, the resistance proportions used in the calculations may be skewed due to variations in blood culture sampling practices. The authors also point out that the report extrapolates data from bloodstream infections to other types of infections without sufficient justification, leading to potential overestimation of the burden of AMR. Furthermore, the attributable mortality estimates used in the report are based on crude data and do not account for important factors such as timing of appropriate therapy and competing outcomes. The authors conclude that more comprehensive, population-based surveillance data are needed to improve AMR control measures and that the "10 million" estimate should be treated with caution until more reliable data are available. They also note that while the UN has called for stronger political commitment to address AMR, there is a need for more accurate and transparent data to inform effective action.The article questions the reliability of the estimate that 10 million people will die annually from antimicrobial resistance (AMR) by 2050, as presented in the UK government's "Antimicrobial Resistance: Tackling a crisis for the health and wealth of nations" report. The authors argue that the estimates are based on flawed assumptions and lack transparency, making them unreliable. They highlight several issues, including the use of data from a non-representative surveillance network (EARS-Net), which only includes data from tertiary hospitals, and the extrapolation of these data to the entire population. Additionally, the resistance proportions used in the calculations may be skewed due to variations in blood culture sampling practices. The authors also point out that the report extrapolates data from bloodstream infections to other types of infections without sufficient justification, leading to potential overestimation of the burden of AMR. Furthermore, the attributable mortality estimates used in the report are based on crude data and do not account for important factors such as timing of appropriate therapy and competing outcomes. The authors conclude that more comprehensive, population-based surveillance data are needed to improve AMR control measures and that the "10 million" estimate should be treated with caution until more reliable data are available. They also note that while the UN has called for stronger political commitment to address AMR, there is a need for more accurate and transparent data to inform effective action.