A quick guide to Generative AI (GenAI)

In Five key factors blocking widespread AI implementation in organizations, we shared the highlights of our Service Management Unlocked (SMU) webinar. This article outlines various terms and concepts associated with artificial intelligence (AI) referenced during the webinar and blog post summary.

Read other blog posts in this series:

>> A quick guide to Supervised Machine Learning

>> A quick guide to Retrieval Augmented Generation (RAG)

———

Google Search’s new ‘AI Mode’ lets users ask complex, multi-part questions

That was the headline of a March 5, 2025, TechCrunch article. 

AI Mode is Google’s new, experimental version of its search engine that removes its traditional Search Engine Results Page (SERP) format of 10 blue hyperlinks and replaces them with an AI-generated summary.

Big news? 

I think so. The 10 blue hyperlink format has been a staple in Google search results since … 1998 — the year it launched.

‍So, yes, the world’s fourth-largest company (by market cap) is making a considerable change to search results 27 years after its inception.

Why do I share this story with you?

To prove, once again, that everyone is talking about — and taking action as it relates to — AI. It’s inescapable. It’s inevitable. 

As stated in A quick guide to Retrieval Augmented Generation (RAG), AI is an essential part of what we do at Xurrent, showing up in the features and functionality of our ITSM platform. As a result of our investment in secure AI, over 90% of Xurrent’s customers have already enabled AI and are realizing incredible productivity gains.

We are committed to creating authentic AI products that make a real impact, tools that are sophisticated yet easy to understand.

As part of understanding all things AI, it’s essential to be well dialed in on definitions you’ll be sure to hear repeatedly. Generative AI (GenAI) is one of them.

This article defines GenAI and shares how we currently leverage it in the Xurrent platform.

What is Generative AI (GenAI), as defined by … AI 

I can’t think of a better way to define GenAI than by asking a few AI assistants. 

I entered the following prompt into Gemini, ChatGPT, and Claude: 

‍In 2 sentences, please define Generative AI.

I then took those three definitions, went back to Claude, and entered the following prompt:

‍Combine the following three definitions of GenAI into a single, two-sentence definition.

Here is the “final” definition of GenAI: 

‍Generative AI refers to artificial intelligence systems that create new content — such as text, images, music, code, or audio — by learning patterns from existing data and applying those patterns to produce original outputs. These systems use sophisticated models like neural networks to generate novel content that mimics human-created work while introducing unique combinations and variations not explicitly programmed into them.

Pretty solid. And a bit meta.

What is the relationship between GenAI and Retrieval-Augmented Generation (RAG), Supervised Machine Learning, and Large Language Models (LLMs)?

GenAI can serve as the generative engine in more complex architectures, such as RAG, as discussed in this article. In RAG systems, GenAI’s creative capabilities are enhanced by retrieval mechanisms that pull relevant information from external knowledge bases.

Supervised Machine Learning, as we outlined here, excels at making precise predictions based on labeled examples. GenAI creates new content from patterns in vast datasets.

The power resides in the combination: using Supervised Machine Learning to classify inputs and trigger appropriate GenAI responses. This partnership enables systems to analyze data with statistical precision and then communicate those results through natural, human-like outputs.

LLMs like OpenAI's ChatGPT, Anthropic's Claude, and Google's Gemini, among others, represent sophisticated implementations of GenAI principles. These systems demonstrate GenAI's foundational capabilities in ractical AI assistants and tool.>> A quick guide to Retrieval Augmented Generation (RAG)

Read other blog posts in this series:

>> A quick guide to Supervised Machine Learning

While each platform has unique strengths — Claude excels in nuanced reasoning, GPT in creative tasks, Gemini in multimodal processing, and Copilot in coding assistance — they all embody GenAI's core ability ... to generate contextually relevant content from learned patterns. 

How Xurrent uses GenAI …

In short, we leverage GenAI across a significant portion of the Xurrent platform.

Here is a sampling: 

‍AI-powered Virtual Agent: The Xurent Virtual Agent interacts with end users to assist them as they seek help. The goal is to “bias the Virtual Agent towards action.”

Example: A user enters, “I need a new Oracle database by the end of the year” into the chatbot. The Virtual Agent immediately opens a ticket and routes it using the “AI auto-classifier.” 

The Virtual Agent can … highlight calls to action, summarize and return a knowledge article, register a new request, remind the user that a request may already exist, return details related to degraded services, and more.

AI Request Summaries: All requests in Xurrent are automatically summarized once they reach predefined thresholds. On average, over 2,000 summaries are created per customer per week. This translates to significant productivity gains, which leads to fewer resources being required to support growth demands.

‍AI Summary for “Waiting for Customer” Follow-Ups: If a request remains in the “Waiting for Customer” status for a specified period with no response from the requester, a follow-up message is sent. The system triggers the AI to add a brief summary of the request and specify the required action or information from the requester, allowing the specialist to proceed..

‍AI Knowledge Article Creation and Improvement: The solution to a request or problem is often not written in a single knowledge article or note. Xurrent’s GenAI assists in creating new knowledge articles (KAs) or updating (improving) existing articles to enhance their effectiveness, based on all the notes from that record. It can suggest a subject, description, and instructions for the desired knowledge article … automatically. 

‍AI Automation Builder: Xurrent AI will automatically build a new automation rule based on a prompt. Simply describe in words how the automation rule should work, and Xurrent will automatically build it. This can save 5-30 minutes per automation. That adds up to massive productivity gains over the lifespan of Xurrent usage.

Stay tuned for more on the when, where, and how of Generative AI (GenAI) at Xurrent.

‍P.S. Speaking of AI, did you know … over 90% of Xurrent customers have already enabled AI and are seeing incredible productivity gains? ‍

‍BONUS: Jim Hirschauer, Xurrent’s Head of Product Marketing, recently used Xurrent (yes, we use our own platform!) to create an automation using our AI builder. It took him less than 5 minutes … as opposed to what would have been an hour+ — a massive 90% reduction in time spent building an automation. Not too shabby.