In a recent video on Help Net Security, Ivan Novikov, CEO of Wallarm, delved into the findings of the 2025 API ThreatStats Report. The report shed light on the increasing vulnerability of APIs as the primary target for cyber attacks, particularly due to the growing risks associated with AI technology.
The researchers behind the report uncovered a staggering 1,025% surge in AI-related Common Vulnerabilities and Exposures (CVEs) compared to the previous year, with a total of 439 identified vulnerabilities. What’s even more concerning is that 99% of these vulnerabilities were directly related to APIs, showcasing the heightened risk associated with these interfaces. The vulnerabilities ranged from injection flaws to misconfigurations, as well as emerging memory corruption issues brought on by AI’s reliance on high-performance binary APIs.
To combat the growing threat landscape fueled by AI technology, Wallarm introduced a new category in their ThreatStats Top 10 list: Memory Corruption and Overflow. This category specifically focuses on vulnerabilities stemming from improper memory handling and access, which could result in detrimental security breaches such as unauthorized data exposure, system crashes, and arbitrary code execution. The analysis also emphasized the heightened susceptibility of APIs to buffer overflows and integer overflows when interacting with AI workloads and hardware, underscoring the urgent need for more robust security measures.
As organizations continue to embrace AI technologies to drive innovation and efficiency, it is crucial for them to prioritize the security of their APIs. With APIs becoming the primary attack surface for cybercriminals, implementing proactive security measures is imperative to safeguard sensitive data and mitigate the risk of potential breaches. By addressing vulnerabilities such as memory corruption and overflow in APIs, businesses can bolster their overall cybersecurity posture and protect against evolving threats in the digital landscape.