This paper introduces the novel concept of Time-Traveling Turing Machines (TTTMs) — Turing Machines with the ability to read symbols that have been explicitly sent from their future computational state. We delve into their properties and behavior, particularly emphasizing the importance of maintaining a self-consistent computational timeline. We present theorems demonstrating the universality of these machines and the challenges in achieving faster computations than traditional Turing Machines.