Critical AI Inference your CIO can Trust
summary
Claims a practical ops layer called XTOPS that adds explainability, traceability and dynamic guardrails to MLOps and cuts incident turnaround from months to days. Shows a GuardHat case where GPS errors create 70% false positives that train users to ignore alerts, then builds telemetry, human-in-loop workflows and digital signing so similar failures resolve in 7 days instead of eight months, and estimates risk at ~$2.5M per site with $500k saved per site per year.
AI-written summary · about
chapters (13)
- 0:00Introduction and background
- 1:05AI adoption and governance gap
- 2:55Pillars of trustworthy AI
- 4:17Explainability and adaptive control
- 6:01From MLOps to XTOPS
- 8:07Key differences in XTOPS
- 9:31Measuring trust with metrics
- 11:57Why trust metrics matter
- 12:19Case study introduction
- 13:20Worker safety AI challenges
- 15:50Convincing CIOs with ROI
- 17:14Outcomes and human oversight
- 18:18Q&A on reputational damage
related talks
19mFrom Alerts to Action: AI-Powered Incident Response Systems
27mAI-Powered Reliability: Stop Fighting Fires, Prevent Them
46mCan you trust your AI?
29mAI Conversation
25mAI-Powered O11y in Microservices Architectures
26mSponsored Session: Navigating Open Source AI with Guardrails
17mHarnessing AI for Autonomous Threat Defense in Multi-Cloud Security
33mHuman Factors in the Age of AI Ops: Re-Engineering Trust between Humans and...