It is helpful to divide the global HIV response into three phases: The first, from about 1980 to 2000, represents “Calamity”. The second, from roughly 2000 to 2015 represents “Hope.” The third, from 2015, is unfolding and may be termed “Choices” – and these choices may be severely constrained by COVID, so “Constrained Choices in an era of COVID” may prove more apt. As we take stock of HIV at 40, there are positive lessons for the wider health response — and challenging reflections for the wider impact of the global HIV response. The positive lessons include: (1) the importance of activism; (2) the role of scientific progress and innovation; (3) the impact of evidence in concentrating resources on proven approaches; (4) the importance of surveillance to understanding transmission dynamics; (5) the use of epidemic intelligence to guide precision implementation; (6) the focus on implementation cascades (diagnosis, linkage, adherence, disease suppression); and finally (7) an overarching execution and results focus. Given this remarkable legacy, it seems churlish to ask whether the HIV response could have achieved more. Yet, consider these approximate figures. Development assistance for HIV totals about 100 billion dollars, 70 billion from the USA matched by roughly 100 billion in domestic resources. For 200 billion dollars, should we not have achieved more than 23 million people initiating treatment (very crudely, 10 000 dollars per person on treatment)? Much of the hundred billion dollars of development assistance (roughly half) focused on about a dozen priority countries in eastern and southern African. The larger PEPFAR recipients, with populations of roughly 50 million, each received 5 billion dollars or more cumulatively. And there are further Global Fund contributions of an additional billion dollars in many of these countries. For 6 billion dollars per country, should we have expected more? The World Bank Human Capital Project posits that to maximize human capital formation, countries must ensure that their children survive, are well nourished and stimulated, learn skills and live long, productive lives. Using the Human Capital Index (a composite index based on these factors), South Africa — the largest HIV financing recipient — ranks 126th of 157 countries, below Haiti, Ghana, the Congo Republic, Senegal and Benin. Consider how many recipients of major HIV development finance fall into the bottom fifth: Namibia, Botswana, Eswatini (formerly Swaziland), Malawi, South Africa, Tanzania, Zambia, Uganda, Lesotho, Ethiopia, Mozambique, Cote D’Ivoire and Nigeria. Of course, causality is unresolved and there are several possible explanations: (1) low human capital formation may increase HIV transmission; (2) the HIV epidemic may have intergenerational impacts; (3) the all-consuming focus on HIV may have displaced other health, education and development priorities. Yet, it remains hard to see these data and to argue that successful HIV responses among the largest HIV financing recipients strengthened their wider health sector and human development outcomes. A plausible principle emerges. Narrowly targeted disease-specific emergency responses may lead to disease-specific gains but do not improve governance or national systems capacity or wider disease or development outcomes. This is not to undermine the emergency origins of the HIV response; 2021 is not 2000 and it is unlikely that we would have 23 million people initiating treatment without an emergency response. Yet, there are reasons (intensified by COVID), to suggest that we must pivot towards long-term, integrated, developmental, nationally owned and financed, systems-orientated responses — particularly when both development assistance and national budgets are likely to be constrained in an era of COVID.