Dave Wright, ServiceNow | ServiceNow Knowledge17
summary
ServiceNow describes an Intelligent Automation Engine that adds time‑series predictive analytics, anomaly detection and instance‑specific machine learning for auto‑categorization and risk scoring, with 82% accuracy out of the box and models trained per customer. It requires about 50–100,000 records to reach reliable accuracy, returns “I don’t know” below confidence thresholds, and chooses ARIMA/STL/projection per dataset. Predictive benchmarking expands to 16 metrics and capabilities roll out across Jakarta and Kingston releases.
AI-written summary · about
chapters (11)
- 0:00Introduction and welcome
- 0:47Defining intelligent automation
- 2:11Platform and product releases
- 3:34Predictive failure examples
- 6:22Auto-categorization challenges
- 8:30Mathematical models in ML
- 10:17Data privacy in AI models
- 11:39Benchmarking strategy
- 13:25Predictive to prescriptive analytics
- 15:29AI impact on jobs
- 18:38Closing remarks and future discussion
