This article at SEPP implies that the IPCC generated assumptions to get the results they wanted:
"If what they say is true, then the methodology used is untenable and hence the results are the same. I can't believe anyone would start with an emissions projection, then make an estimate of the relationship between emissions and growth, and finally calculate growth residually. This is the relationship absolutely backward."
But that, apparently, is exactly what the IPCC's economic project did: It worked backward. Formally known as the Special Report on Emissions Scenarios, the official IPCC group set about creating economic forecasts that appear to support average world temperature increases of somewhere between 1.4 and 5.8 degrees over the next 100 years. But the SRES, by imposing emissions forecasts on economic models, adopted scenarios that economists say make no sense.
This article at SEPP implies that the IPCC generated assumptions to get the results they wanted:
It appears the IPCC apparently is not above a few ad-hoc changes to their reports as well:
http://www.sepp.org/keyissue.html
The IPCC Controversy: In May 1996, unannounced and possibly unauthorized changes to the latest United Nations report on climate change touched off a firestorm of controversy within the scientific community. The Intergovernmental Panel on Climate Change (IPCC), the science group that advises the United Nations on the global warming issue, presented the draft of its most recent report in December 1995, and it was approved by the delegations. When the printed report appeared in May 1996, however, it was discovered that substantial changes and deletions had been made to the body of the report to make it "conform to the Policymakers Summary." The clandestine changes put a spin on the report's conclusions that "the balance of evidence suggests a discernible human influence on global climate." Lead authors of the crucial--and doctored--Chapter 8, dealing with the detection and attribution of climate change, have since backed off from this conclusion and now admit that it may take 10 years or more before any human influence on climate can be detected. For commentary and letters on this issue, see IPCC.