Runtime AI brings AI-powered capabilities directly into the execution of a Nintex Workflow or App. Unlike design-time AI, which assists with building workflows, Runtime AI enhances how your live workflows operate by analyzing content, generating summaries, detecting entities, and making light, context-aware decisions as the workflow runs.
Runtime AI Actions are lightweight, single-step automations that use large language models (LLMs) such as GPT and services like Azure Document Intelligence to perform specific content generation or analysis tasks. They are easy to configure and fit naturally into workflow execution, allowing you to add intelligence without complexity.
All actions are delivered through the Runtime AI Platform Service and are accessible via API for integration across Nintex products. In this release, there are 11 out-of-the-box actions available in Nintex Workflow.
Videos
- Analyze sentiment
- Detect language
- Extract information from document or image
- Extract text from document or image
- Summarize text
- Categorize document
- Detect PII
- Extract information from text
- Summarize document
- Translate Text
- Categorize Text
Consistent Output Model
Every AI Action returns an Object variable. Even simple actions like Analyze Sentiment return a single value within an object. This consistent structure lets you adopt new output fields over time without changing the action’s output format, simplifying downstream logic and future enhancements.
Analyze Sentiment
What it does
Automatically detects the sentiment of a given text to support intelligent branching, escalation, or prioritization in workflows.
Why it matters
Understanding sentiment helps teams gauge customer satisfaction, employee morale, and public perception without manual review.
Typical use cases
-
Social media monitoring: Track sentiment across mentions. Spikes in negative sentiment can trigger PR outreach or product follow-up.
-
Customer support feedback analysis: Analyze post-chat surveys. Negative results trigger a follow-up workflow within 24 hours.
Detect Language
What it does
Detects the primary natural language of a given text to enable dynamic routing or customization of downstream processes.
Why it matters
Language detection is foundational for routing, translation, and personalization in multilingual environments.
Typical use cases
-
Multilingual customer support routing
-
Localization workflows for user-generated content
Extract Text From Document or Image
What it does
Uses text recognition to extract all handwritten and printed text from a document or image. Ideal for feeding into other AI actions like summarization or sentiment analysis.
Why it matters
Automates manual data entry and unlocks insights from scanned or photographed content.
Typical use cases
- Field agent reports based on photos of handwritten forms
- Archiving historical documents so they are searchable and indexable
Extract Information From Doc or Image
What it does
Extracts specific, structured information from a given document.
Why it matters
Enables intelligent automation when you need data points, not just raw text.
Typical use cases
-
Record storing: Extract invoice details and route based on thresholds
-
Customs declaration scanning: Extract declared items and values for risk flagging
Summarize Text
What it does
Generates a concise summary from a block of text, guided by user-provided instructions.
Why it matters
Speeds up understanding of long content for faster decision-making.
Typical use cases
-
Support ticket summarization for quicker triage
-
Meeting notes digest into key decisions and action items
Translate Text
What it does
Translates text from one language to another.
Why it matters
Enables global communication, localization, and accessibility.
Typical use cases
-
Multilingual customer support
-
Global form or survey translation and analysis
Categorize Document
What it does
Analyzes an entire document (text, structure, metadata) and assigns a document type or category.
Why it matters
Lets you identify document types before deciding on routing or processing steps.
Typical use cases
- Healthcare document intake with routing to billing, diagnostics, or pharmacy
Categorize Text
What it does
Automatically classifies incoming or generated text into predefined categories.
Why it matters
Enables intelligent routing and handling of content at scale without manual triage.
Typical use cases
-
Customer support topic routing
-
HR feedback processing and reporting
Detect PII
What it does
Identifies and flags personally identifiable information (PII) in text, including names, email addresses, phone numbers, ID numbers, and credit card details.
Why it matters
Reduces compliance risk and manual review by automatically detecting sensitive data in unstructured content.
Typical use cases
-
HR compliance audit before external sharing
-
Contract review to identify personal data prior to publishing or archiving
Extract Information From Text
What it does
Extracts structured data from unstructured text such as forms, transcripts, or messages.
Why it matters
Replaces manual parsing and brittle custom logic with scalable extraction.
Typical use cases
-
Finance automation after speech-to-text
-
Insurance claims intake from written incident reports
Summarize Document
What it does
Generates concise summaries of long or unstructured documents.
Why it matters
Supports faster understanding and decision-making without reading the entire document.
Typical use cases
-
Contract review for key terms, obligations, and dates
-
Customer feedback analysis for themes or concerns
-
Insurance claim intake summaries
Runtime AI Actions extend Nintex Workflow with practical, execution-time intelligence that fits naturally into existing automation patterns. With these 11 out-of-the-box actions, workflows can analyze content, extract meaning, and respond dynamically as they run, without adding complexity to design or maintenance.
As part of the Runtime AI Platform Service, these actions provide a consistent, extensible foundation for building smarter workflows today while leaving room to grow as new AI capabilities are introduced.