Akamai’s latest research reveals that AI-powered applications in the Asia Pacific and Japan (APJ) region were targeted by 51 billion web attacks in the past year alone, representing a 73% year-over-year increase.
Increasing web attacks
Attacks in the Financial Services sector (27 billion incidents) and the Commerce sector (18 billion incidents) underscore the significant widening of the attack surface across industries due to the rapid adoption of AI technologies.
Akamai's API Security Impact Study has revealed that the average estimated cost of API security incidents exceeds US$580,000, which should be allocated for investments in AI and tech infrastructure.
It also found that AI is reshaping security operations by enhancing detection and response capabilities. However, the technology opens applications to new vulnerabilities, highlighting a greater need for modernised AI-aware security solutions.
Firewall for AI
Akamai recently unveiled new tools designed to secure AI workloads: Firewall for AI and API LLM Discovery. These innovations claim to provide a multilayered protection against adversarial queries and data scraping, and automatically discover and manage LLM and GenAI API endpoints.
Firewall for AI claims to include the following features:
- Detects and blocks prompt injection, jailbreaks, and adversarial inputs before reaching your LLM
- Response filtering and moderation
- Multilayered input and output guardrails to prevent sensitive data exposure
- Real-time AI monitoring
- Enterprise-scale security
- Flexible deployment options
- Policy-based controls
