Safeguard Your APIs: Machine Learning Solutions
summary
Claims machine learning is required to scale API security as enterprises report an average of 15,000+ production APIs, 76% having seen breaches and APIs growing rapidly (examples cite ~200% growth). The solution wires IBM API Connect + DataPower to NoName Security ML engine via out‑of‑band mirrored traffic, sends full request/response for learning, returns automatic blocking rules to DataPower, and is deployable SaaS or on‑prem. Demo discovers 120 APIs across 35 hosts, detects a credential‑traversal attack and enforces a block successfully.
AI-written summary · about
chapters (20)
- 0:00Introduction
- 0:20API Security
- 1:11Digital Journey
- 2:30IBM API Connect
- 3:08Datapower Gateway
- 3:37API Flow
- 5:39NoName Security
- 7:10Architecture
- 8:25Knowin perspective
- 13:46Deployment
- 15:08Demo
- 19:14Query catalog
- 20:01Detect changes
- 20:47Data types
- 22:12Workflows
- 23:06Consumers
- 24:44Security
- 25:53Runtime Security
- 30:35Management API
- 31:29Outro
related talks
23mMachine Learning powered API governance
24mAPI security, mitigate API abuses using machine learning
05mSecuring the Future of APIs: AI-Powered Threat Detection & DevSecOps Integration
12mSweet or Heat
33mImproving API Security Posture.
24mSecuring the Modern API Ecosystem
55mApisecure, 14 & 15 March 2023 - Sagar Bhure, Security Engineer at F5.
19mAvoid These API Security Pitfalls: Expert Insights