All $447 trillion worth of global financial assets are priced relative to the federal funds rate, the interest rate at which banks and credit unions lend reserves to each other overnight. The ability to more accurately forecast this rate can enhance long term economic growth and allow for better government, corporate, and personal financial planning. Currently, the most accurate forecast of the federal funds rate is the federal funds futures market, financial contracts that represent market opinion of where the daily federal funds rate will be at the contract's expiration date. Since year 2005, Chicago Mercantile Exchange (CME) has been listing 30-day futures 24 months out. Starting in 2013, CME has been offering 30-day futures 36 months out. It is my research hypothesis that I may be able to predict the federal funds rate better than the market at least on the later half of the 24-month and 36-month prediction periods. The objective of this research is to forecast future federal funds target rates using Artificial Intelligence (AI). Time series training data of economic indicators from Federal Reserve Economic Data (FRED), maintained by the Research division of the Federal Reserve Bank of St. Louis, was used to train the neural network model. The combination of a Multilayer Perceptron classifier and a novel dynamic decision tree model was used to generate the predictions. The time-series decision tree model was mainly used to prevent jittering in the forecasted rates. The selection of attributes, especially the selection and decision on significant overlay data (the projected values of the economic indicators during the prediction period), was experimented to see how it may influence the outcome of the prediction. The hybrid of a decision tree and MLP's forecasts matched (the Tukey Honest Significance p value > 0.05) the federal funds rate closely from years 2008 - 2009, 2010 - 2011, 2012 - 2014, and 2015 - 2017, while the predictions of the federal funds futures contract only matched the federal funds rates in 2012 - 2014. The performance analysis shows an average of 66.7 and 65.9% improvements in the deviation from the target federal funds rate and RMSE, respectively. The improvement in the forecast of this model can also save a significant amount of money for governments, corporations, and individuals. The model predicts the federal funds rate more accurately by 0.5% or 50 basis points, which can save the federal government 105 billion dollars annually. With a 50 basis point return, BlackRock (the world's largest asset management company), can save 31 billion 400 million dollars per year. An individual who resides in the zip code 95030 can earn 1.5 million dollars over 30 years.