Harvard historian Naomi Oreskes, Princeton geoscientist Michael Oppenheimer and some other "scholars" maintain that climate scientist actually have underpredicted evidence of human caused global warming:
In fact, it increasingly appears that, if there is any systemic bias in the climate models, it’s that they understate the gravity of the situation. In an interesting paper that appeared in the journal Global Environmental Change, a group of scholars, including Naomi Oreskes, a historian of science at Harvard, and Michael Oppenheimer, a geoscientist at Princeton, note that so-called climate skeptics frequently accuse climate scientists of “alarmism” and “overreacting to evidence of human impacts on the climate system.” But, when you actually measure the predictions that climate scientists have made against observations of how the climate has already changed, you find the exact opposite: a pattern “of under- rather than over-prediction” emerges. The scholars attribute this bias to the norms of scientific discourse: “The scientific values of rationality, dispassion, and self-restraint tend to lead scientists to demand greater levels of evidence in support of surprising, dramatic, or alarming conclusions.” They call this tendency “erring on the side of least drama,” or E.S.L.D. for short.
Alarmists Oreskes and Oppenheimer must be joking. Or do they think that climate scientists have underpredicted the fact that there has been no global warming during the last almost 18 years?