Any weather or climate forecast is a statistically-based guess.  Anything beyond about three days:  a month, three months, a year, in 2030, or in 2100 is a shot in the dark.  The longer-term the forecast, the increasingly less accurate it will be.  Call it weather or climate, the calculations and issues are all the same.

Just as problematic, (1) The parameters used in any computer modeling can be set up to predict any outcome desired (i.e., due to personal bias or that of the agency financing the forecast), and (2) seemingly-insignificant changes to any model's parameters can result in a radically-different forecast.  Computer forecast models of the weather, like those of the stock-market, have no validity; they are only accurate to the degree that the future is just like the past, which is never true.

See these various issues affecting the accuracy of weather forecasting.