Agentic and generative AI features overview for Pega Smart Investigate Agentic Automation
Pega Smart Investigate Agentic Automation™ (SIAA) introduces a set of features that transform payment investigation operations by shifting from manual, investigator-driven tasks to coordinated, governed, and intelligent automation. These features operate within prescribed procedures and align with internal policies, industry standards, and regulatory requirements. This design supports operational consistency, helps improve quality, and reduces risk across high-volume financial investigation processes.
Agentic features
SIAA is focused on supporting the six key agentic features. These features provide a structured, orchestrated approach to common investigation tasks:
- Smart Investigation Agent: Checks whether a payment was completed and paid correctly. Can call Swift services for payment status, perform "stare and compare" across instructions, and detect anomalies.
- Accounting Agent: Completes iterations of ad hoc adjustments autonomously and advises users on how to complete adjustments.
- Correspondence Agent: Drafts Swift or email correspondence and advises on the best way to correspond with parties based on the Case and counterparty preferences.
- Compensation Agent: Completes calculations and models alternative calculations based on case data and diverse inputs. Triggers compensation resolutions.
- ProcessAI 'Next Best' Resolution activity: Adaptive and Predictive Models suggest the next best resolution activity to perform to complete an investigation.
- Agentic Email: Generative AI-powered email handles various inbound email demands, including:
- Queries on transactions
- Requests for payment status
- Advice on compensation
- Answers to common questions, for example, charge issues
Generative AI features
SIAA includes two generative AI features through Pega GenAI Coach™. These features help reduce cognitive load, accelerate Case orientation, and democratize technical expertise across investigation teams:
- Case Summarization: Synthesizes extensive Case information, including attachments, audit updates, notes, correspondence, and life cycle events instantly, to support the reduction in Case orientation time.
- Message Summarization: Translates complex Swift ISO 20022 XML format and MT format messages into natural language summaries, which removes the technical barrier that traditionally restricted payment exception management to specialized back-office teams.
This Topic is available in the following Module:
Want to help us improve this content?