The Python Testbed for Federated Learning Algorithms (PTB-FLA) is a simple Python FL framework that is targeting edge systems and is by its design easy to use by human ML&AI developers. The original PTB-FLA development paradigm intended for humans consists of the four phases (producing the sequential code, the federated code, the federated code with callbacks, and the PTB-FLA code, respectively), and hence dubbed the four-phases (development) paradigm, was validated in the case study on the logistic regression. In this paper, we adapted the original paradigm into the two new paradigms for ChatGPT, named the adapted four-phases paradigm and the adapted two-phases paradigm, respectively. In tune with its name, the latter consists of two phases (producing the sequential and the PTB-FLA code, respectively). We successfully validated both new paradigms using the same case study on logistic regression that was used for the original paradigm. The results are positive and encouraging as the resulting program codes are of better quality than the codes solely made by humans using the original paradigm.