<Where I have problems is with computer models that because someone with a PhD in the field made them are naievely accepted in our technophiliac society as being the truth. The historical fact is that many flawed climate models have been touted in the press, with enthusiastic support by scientists, as accurately predicting the future. They don't.

Some models have the problem of predicting something that can't be tested and verified against. Models that predict things hundreds of years in the future, for instance. They are harder to get right because you don't have positive feedback to use to find problems. Many climate models are certainly in this category. They should be less trusted than other models. Until we develop warp-drive and start popping back and forth in time whenever an alien sneezes, they should be taken with a grain of salt.

Other relevant examples include the hilarious computer generated solar system models featured in Sagan's books & elsewhere. None of them generated Hot Jupiters or solar systems like we've detected so far. Instead, they gave the PhD astronomers what they expected.

I think this is just a case of the correction process at work. Anyway, for all we know at this point, those old models may be mostly right. Time will tell.