High level automation of ship technology promises to revolutionize maritime activities with benefits like reduced operational costs and enhanced safety. Yet, this technology also encounters significant challenges in cybersecurity. Cyber-attacks on autonomous or highly automated ships could lead to disastrous outcomes, including environmental damage and threats to safety and security. These attacks might compromise key systems like GPS or communication channels, resulting in loss of control and potential accidents. To mitigate these risks, this paper focuses on employing artificial neural networks for accurately estimating a ship’s position using data from various sensors, such as gyroscopes, propeller speed, propeller pitch, weather and sea states, drafts, and engine load. This method may help in detecting and counteracting any manipulated positional data and make cross-verification of GPS data and other speed measurement technologies like Doppler sensors. Additionally, the effectiveness of artificial neural networks in this context is demonstrated by testing the algorithm with new data sets, manipulating the ship’s position to assess its accuracy. The findings indicate that leveraging propulsion system data within dead reckoning systems to develop an artificial neural network could provide a reliable solution for detecting and mitigating GPS spoofing-related cyber threats in the maritime industry.