Global market research firm Gartner predicts that by 2027, over 40% of AI-related data breaches will occur due to the use of generative AI.
According to Gartner’s report on Tuesday, the rapid adoption of generative AI technology is outpacing advancements in data security, leading to a surge in inappropriate cross-border generative AI incidents.
A Gartner VP researcher, Joerg Fritsch, explained, “Unintended cross-border data transfers often occur due to insufficient oversight, particularly when GenAI is integrated into existing products without clear descriptions or announcement. Data breaches can happen when prompts requiring security are sent to APIs in unknown locations.”
Fritsch further noted that the lack of global AI standardization is causing operational inefficiencies. Without consistent global standards addressing AI and data governance, companies must develop strategies tailored to specific countries or regions, which can negatively impact the scalability of global AI operations.
Gartner recommends several strategic measures to mitigate AI data breach risks: expanding data governance frameworks, enhancing AI oversight, forming governance committees, implementing advanced encryption for sensitive data, and purchasing risk and security management (TRiSM) products.
Gartner emphasized that by 2027, AI governance will become mandatory in all sovereign AI laws and regulations worldwide, to which companies need to respond preemptively.