Code injection attacks are becoming more common, which puts software system integrity and confidentiality at risk and presents a serious danger to cybersecurity. Conventional protection systems frequently find it difficult to stay up with the ways that bad actors are changing their strategies. In order to assess and mitigate code injection attacks, this research combines the advantages of block chain technology, artificial intelligence (AI), and recurrent neural networks (RNNs) to provide a novel solution. The proposed system integrates block chain’s immutable and transparent ledger capabilities with AI-driven anomaly detection. Specifically, RNNs are employed to model and predict normal and anomalous code execution patterns, enhancing the detection of sophisticated injection attacks. The block chain framework ensures tamper-proof storage of historical execution data and attack patterns, enabling real-time and retrospective analysis. The paper illustrates the effectiveness of this hybrid strategy in enhancing the precision and resilience of attack detection mechanisms by methodically assessing its performance. Results indicate that the integration of RNN-based AI with block chain technology significantly enhances the system’s ability to identify and respond to code injection threats, providing a promising avenue for advancing cyber security defences.