1、The State of AI andAPI Security:Navigating the Agentic EraTable of Contents1Executive Summary.2Key Findings.5Drivers for AI Adoption.8 API Development Trends.9 API Security Challenges in the Agentic Era.10 Salt Labs Analysis of Customer Data.12 Monitoring and Securing APIs.14Generative AI and API Se

2、curity Risks.17Measuring ROI in API Security.20 Conclusion and Recommendations.21About Salt.23 Executive Summary IThe Emerging Agentic Security GapThe Emerging Agentic Security GapSecurity leaders increasingly recognize that AI systems introduce entirely newSecurity leaders increasingly recognize th

3、at AI systems introduce entirely newoperational risks.Our survey shows that boards and executive leadership are nowoperational risks.Our survey shows that boards and executive leadership are nowactively scrutinizing AI security:actively scrutinizing AI security:79%report increased executive scrutiny

4、 of AI security risks.79%report increased executive scrutiny of AI security risks.69%of boards are concerned about sensitive data leakage through AI prompts69%of boards are concerned about sensitive data leakage through AI promptsor models.or models.39%are specifically worried about autonomous agent

5、s acting without human39%are specifically worried about autonomous agents acting without humanoversightoversight.Despite this boardroom mandate,organizations admit to a severe Confidence GapDespite this boardroom mandate,organizations admit to a severe Confidence Gapwhen relying on traditional secur

6、ity stacks to address AI-driven threats.The visibilitywhen relying on traditional security stacks to address AI-driven threats.The visibilitygap mentioned earlier is compounded by a profound failure in legacy tooling:gap mentioned earlier is compounded by a profound failure in legacy tooling:AboutAb