Convergence of Chaos Engineering and Revolutionized Technology Techniques
summary
Argues that chaos engineering can harden IoT, AI, security and human-augmentation stacks by running scoped failure experiments and validating defenses. Describes a prototype ML pipeline to classify postmortems from historical outages for faster incident triage, and lists concrete chaos experiments: generate slow responses and cascade failures in IoT, intercept integration points, inject latency into security controls, simulate permissions collisions and disable MFA in production-like environments. Recommends game days as a human-centered way to run those experiments and measure resilience.
AI-written summary · about
chapters (14)
- 0:00Start
- 3:04Internet of Things
- 3:57Resilience
- 9:04Data
- 9:52Machine Learning
- 10:26Natural Language Processing
- 13:29Cybersecurity Security and reliability are often hadden
- 14:17Virus Evolution
- 15:04Fuzz Testing
- 15:27Human Factors
- 17:53Human Augmentation
- 18:32Disclaimer: Automation
- 18:54Challenge 1: Reliability
- 19:21Challenge 2: Safety
related talks
32mUnleashing Chaos with GenAI Innovations
24mChaos and Pain in Machine Learning, and the ‘DevOps for ML Manifesto'
26mRunning Secure AI/ML Experiment in a Unified MLOPS-DevSecOps Pipeline
45mHardening AI/ML Systems - The Next Frontier of Cybersecurity
21mMastering the DevOps Machine Learning Pipeline
44mAdvancing Cybersecurity with AI and Machine Learning: Opportunities and Challenges
26mCompleting the machine learning loop
24mTesting Techniques for Critical Deep Learning Applications