As Winston Churchill said:
No one pretends that democracy is perfect or all wise. Indeed, it has been said that democracy is the worst form of government, except for all the others that have been tried from time to time.
Peer review, like science itself, should be self-correcting. A few bad papers may get accepted but eventually the work should be properly evaluated and treated accordingly.
But recently there is a growing trend for pay-to-publish journals to pretend to do peer review, while accepting just about any paper that an author will pay for.
Science has recently published a detailed article on this topic, including an experiment involving an obviously inaccurate article submitted by fictitious authors. The results are not encouraging.
The most basic obligation of a scientific journal is to perform peer review, arXiv founder Ginsparg says. He laments that a large proportion of open-access scientific publishers “clearly are not doing that.” Ensuring that journals honor their obligation is a challenge that the scientific community must rise to. “Journals without quality control are destructive, especially for developing world countries where governments and universities are filling up with people with bogus scientific credentials,” Ginsparg says.