Authored by Minaz Merali and Abhishek Yadav

IBM Z has served as the trusted system of record for high value transactional processing for decades. Within this transactional data are behavioral patterns that can reveal fraud risk, customer intent, operational anomalies and more. Yet many organizations still copy mainframe data to external analytics platforms, introducing unnecessary cost, operational complexity, data latency, governance and compliance risk.

SQL Data Insights Pro eliminates these challenges by bringing AI directly to Db2 for z/OS. Insights become available instantly and securely, without moving data off the platform.

For example, a fraud analyst can begin with a known suspicious transaction and instantly find others exhibiting similar patterns—including signals hidden in payment descriptions without moving data off platform. And a customer analytics team can start with recently churned customers and identify others who show similar behaviors, enabling earlier intervention.

SQL Data Insights Pro delivers this by enabling organizations to analyze structured data—numbers, categories and more—and unstructured text together in place using a single unified model. With familiar SQL, teams can surface records that are similar in meaning, behavior or context—unlocking insights that traditional value‑based queries alone cannot detect.

SQL Data Insights Pro stands apart by bringing:

• Pattern and anomaly discovery: Detect signals, clusters, and unusual activity across large datasets without any risk or cost of moving data or impacting application SLAs.
• Unified structured and unstructured analysis: Unified relational data with notes, descriptions, or logs to uncover richer insights that would otherwise remain hidden.
• Quick time to value, without needing specialized AI skills: Data scientists are not required to build or operate AI pipelines. SQL Data Insights Pro integrates AI capabilities directly into Db2 for z/OS, allowing database professionals and analysts to run AI-powered queries using the skills they already have.
• Risk-free AI model training and incremental model retraining: Model training runs on IFLs, protecting critical Db2 workloads. Support for incremental retraining allows models to refresh efficiently using only new or changed data.

SQL Data Insights Pro opens new possibilities across industries. Here are a few examples teams within organizations can apply today: 

• Customer retention: Identify customers who resemble those who recently canceled a service, helping teams intervene earlier.
• Product recommendation: Find customers whose purchase behavior resembles those who bought a particular product, enabling more targeted cross-sell and upsell opportunities.

• Fraud detection: Start with a known suspicious transaction and instantly find similar ones, including hidden behavioral patterns in text descriptions.
• Operational anomaly detection: Detect unusual systems or operational events by analyzing structured logs together with free-text messages.

• Entity resolution: Resolve duplicate or incomplete customer profiles for unified customer view.

These insights are generated without exporting data, without external AI stacks and without requiring data science expertise.

SQL Data Insights Pro enables meaning-based analysis across all data types without moving or exposing data—reducing architectural complexity, minimizing operational risk, supporting regulatory and compliance requirements and accelerating time to decision.

IBM SQL Data Insights Pro will be generally available on 20 March 2026.

To help teams get started quickly, we will introduce industry‑specific starter kits shortly after GA. These kits include curated datasets, prebuilt AI models, and ready‑to‑use SQL AI queries that teams can adapt to their own data and business scenarios. 

Register for our upcoming webinar