Since the advancements in Artificial Intelligence (AI) around the world, it is a common belief that nothing is impossible due to AI penetration. However, reality is far from this notion when it comes to sign language recognition. Since human sign language incorporates face, facial impressions, hand gestures and dynamic hand movements along with changing gestures and even body movements have made it impossible for improvements around sign language recognition so far. A paper published in 2024 indicated that there has been no progress at all in sign language recognition despite some promising outcomes more than 15 years ago [Premaratne et al. in Communications in Computer and Information Science , 2014 CCIS, pp. 161–168, 2024). Due to the complexity of the sign language recognition problem as stated above due to array of movements of different body parts which limit any system clearly detecting and tracking the gestures, moods and movements for translation, the authors plan to break down the problem into manageable areas that would eventually result in a world-changing outcome. In this research presentation, the authors tested the use of Long Short-Term Memory (LSTM) neural networks approach to handle up to 80 frames associated with each short sign language phrases such as ‘How are you’ and ‘I am fine’ to see whether LSTM type architecture would result in any promising outcomes. The results indicated the limitations of this new RNN systems’ ability to handle sign language.