The research on drug poisoning explores the dangerous consequences of materials on the frame, particularly targeting the liver and kidneys’ failure to take away pills from the bloodstream. Drug toxicity can arise from deliberate overdoses and unintentional poisoning, mainly to a large variety of damaging health consequences, such as paralysis, confusion, seizures, and loss of life. This research explores the factors affecting drug toxicity, along with absorption, distribution, metabolism, and elimination (ADME), along with dose–response relationships, cumulative effects, and individual variability. ADME/Tox properties are crucial in the improvement of safe capsules, and awful ADME/Tox homes often result in drug failure. The studies make use of a laptop version blending cellular automata, gadget getting-to-know, and optimization methods to predict the toxicity levels of five drug molecules. Sample records of the atomic makeup of the drug changed into studies for molecular clues and damage forecasts. This research creates a forecast model for drug poisoning that blends pharmacology statistics with superior laptop methods. By understanding ADME tendencies and different toxicity-related factors, this version tries to enhance drug safety critiques. The studies efficiently created a predictive framework for drug toxicity, giving a tool for measuring the safety profiles of drug combos. Machine getting-to-know and optimization techniques confirmed achievement in danger prediction, making major steps closer to extra dependable and powerful drug evaluation tactics. This research should enhance the version by adding real-world scientific records and widening it to a degree of more diverse substances in future research. Further development with superior neural networks and adaptable learning models may additionally enhance the predicted accuracy, helping in drug finding and regulatory compliance.