Create value from diverse IoT data and initiatives – at the edge, in the cloud or anywhere in between – as you advance toward the artificial intelligence of things (AIoT)

No, streaming AI is not applicable to all industries, but it provides significant value in sectors that deal with real-time data, events or dynamic environments. Its usefulness depends on whether an industry has a continuous data flow, a need for real-time decision making and operational processes that benefit from up-to-the-second insights.

• Banking & finance: Fraud detection, real-time trading analysis, compliance monitoring.
• Manufacturing: Predictive maintenance, process control, quality assurance.
• Retail & e-commerce: Dynamic pricing, customer behavior tracking, real-time inventory updates.
• Telecommunications: Network optimization, outage detection, customer service routing.
• Transportation & logistics: Fleet tracking, route optimization, traffic prediction.
• Health care: Patient monitoring, anomaly detection in medical devices, emergency response.
• Energy & utilities: Smart grid management, load forecasting, fault detection.
• Public sector & defense: Border monitoring, cyber threat detection, emergency services dispatch.

Computer vision enables machines to interpret and act on visual data. The top 10 use cases span various industries and include:

1. Quality control: Automated defect detection in manufacturing.
2. Security: Facial recognition and intrusion detection.
3. Health care: Medical imaging analysis for diagnostics.
4. Retail: Monitoring inventory and analyzing customer behavior.
5. Autonomous vehicles: Object and lane detection.
6. Agriculture: Crop and livestock monitoring.
7. Logistics: Package tracking and damage detection.
8. Smart cities: Traffic monitoring and waste management.
9. Workplace safety: PPE compliance and hazard detection.
10. Document processing: Optical character recognition (OCR).

Edge AI is a specific type of edge computing focused on running AI models directly on edge devices. Edge computing is a broader concept that involves processing any type of data near its source to reduce latency and bandwidth usage.

Yes, SAS Event Stream Processing is a streaming data platform, but more specifically, it is a real-time analytics engine. It's designed to ingest, process and analyze high-velocity data streams to detect patterns or anomalies for low-latency decision making.

SAS IoT analytics supports agentic AI by providing the foundational capabilities needed for autonomous, goal-driven systems. It enables AI agents to:

• Sense the environment via real-time data ingestion.
• Think by applying analytics and models.
• Act through edge deployment and process control.
• Adapt using continuous learning and feedback loops.

GenAI is a broad category of AI that creates new content. LLMs are a type of GenAI specifically for text, while RAG is an architecture that enhances LLMs by adding external, real-time knowledge to improve accuracy.

Yes, SAS supports the creation and use of digital twins by providing the data, analytics and machine learning infrastructure to power them. While SAS doesn't offer a 3D modeling platform, it delivers the real-time data ingestion, streaming analytics and predictive modeling needed to simulate and optimize physical assets.

• Manufacturing: Simulate equipment wear and performance under different operating conditions.
• Utilities: Model energy usage and grid behavior to optimize load balancing.
• Transportation: Monitor the condition of vehicles and simulate route or maintenance impact.
• Smart cities: Mirror traffic systems or water infrastructure for scenario planning.