MORE THAN six months have passed since the World Health Organization and its allies launched the organization's "3 by 5" initiative, an effort to get antiretroviral medicines to 3 million HIV-infected people by the end of 2005. That target is a long way from where the world is at the moment. At the end of 2003, only 400,000 patients were receiving treatment, and the prospects for a rapid expansion were clouded by the Bush administration's attitude toward cheap generic drugs. On Sunday, however, the administration announced a promising shift in policy. If it is implemented effectively, there will soon be a big jump in the number of patients receiving treatment in the 14 African and Caribbean countries covered by President Bush's AIDS initiative.

Previously the administration had instructed the medical charities and other agencies that implement its program to use branded AIDS medicines, even though these are expensive and unavailable in simple pills combining three drugs in one capsule, and even though some target countries (Tanzania, for example) lean strongly toward using combination pills. Moreover, the administration refused to accept assurances from WHO that the combination pills offered by Indian generic manufacturers are safe and effective, while also refusing to say what was wrong with the WHO certification process or what alternative process it might prefer.

Now the administration has said that generic drugs may be acceptable, and it invites the generic manufacturers to submit their products to the Food and Drug Administration so that they can be certified for use in U.S. programs in developing countries. Critics complain that the WHO certification process is still not getting due respect, and there's something to these objections. But the Bush administration claims the certification process will take a maximum of eight weeks -- speedy by FDA standards.

Meanwhile, the administration's announcement is also intended to encourage branded manufacturers to create combination drugs that compete with the generic companies. This, too, is welcome: The objective is to draw in as many potential suppliers as possible, so that doctors have a choice of therapies to fight the virus in different contexts and to cope with resistance as it inevitably evolves. The challenge of treating AIDS is unfortunately so huge -- about 35 million people live with the virus, and there are about 5 million new infections annually -- that there is room for multiple companies in this market. If the world succeeds in meeting the "3 by 5" targets, there will likely be more targets afterward: "6 by 8," "10 by 10" and, horrifically, so on.