How ServiceNow is using generative AI capabilities across its operations

As the emerging technologies team within ServiceNow digital transformation (DT), we’re always searching for new technology that can solve complex problems. When we saw how generative AI capabilities could deliver rich experiences and unlock the potential of our technology, people, and processes across the organization, we immediately pivoted to embrace it.

This year alone, we will have brought more than 15 Now Assist use cases to production on the Now Platform—many of which are generally available in the Vancouver release, including Now Assist for Customer Service Management, IT Service Management, HR Service Delivery, and Creator.

These innovations build on our solid foundation of AI. One of the DT organization’s first forays into AI was in 2017, when we looked for solutions to make our IT service desk more efficient. We used the machine learning capabilities in Predictive Intelligence to help predict the business service and route incidents to the proper assignment groups.

That use case is still in place today. Since then, we’ve embedded AI across more platform experiences, from calculating our self-service rate to using knowledge base articles for case deflection, recommending learning courses to our employees, and using AI Search across our operations.

Embracing new platform capabilities

As customer zero, ServiceNow embraces new platform capabilities well before our customers do. We’re leaning into two capabilities to embed generative AI into our workflows:

1. Summarization – overviews and insights to help employees reduce manual work
2. Content generation – creation of intelligent search results, work notes, and knowledge base articles to help employees find and digest information more rapidly

Here are some of the results we’re seeing as we move our use cases from pilots to production:

• AI Search with GenAI ​– This generates actionable answers to user-asked questions to significantly improve self-service. Data is sent to a large language model (LLM), which summarizes the information into a single answer. Follow-up questions and answers are supported. This differs from simple AI search, where a list of sources is generated for the user to sort through. Early results show positive feedback from employees.
• Now Assist for IT Service Management – This is an internally developed use case to increase agent productivity. It uses generative AI’s summarization features to produce a summary of incident work notes, saving time and effort. In our initial results, we’ve seen increased agent satisfaction.
• Now Assist for Creator (formerly text-to-code) – Developers often face repetitive and time-consuming work when creating the same code for routine commands. Our developers can now write plain, natural-language text descriptions of the type of code they want. Generative AI converts the text into intelligent code suggestions or complete code, which can be shared, reviewed, edited, and executed. This can lead to a boost in developer productivity and development cycles.
• Generative AI Controller ​– The controller allows us to connect ServiceNow instances to our domain-specific LLMs or general-purpose LLMs such as OpenAI and Azure OpenAI. We can easily embed generative AI in any workflow without recreating LLM integrations from scratch, giving us limitless opportunities to use generative AI. We’re using it for knowledge base content generation.

We’ve also used generative AI to help develop custom, function-specific solutions. A question-and-answer capability helps our sales team find answers to specific questions about our products and solutions, including source citations. Although this capability is in early adoption, we’ve seen improved productivity and user experience.

Reaping the rewards

Summarization and code generation capabilities offer three major benefits in these early stages of adoption:

• Accelerated productivity by automating mundane work for our agents, administrators, employees, and developers so that they can focus on meaningful work
• Increased agility by adapting quickly with continuous updates, new use cases, and choice of LLMs to drive greater visibility and control
• Transformed experiences by delivering fast, personalized, and trusted resolutions to employee issues

As we embrace generative AI to modify old workflows, generate new workflows, and enhance employee experiences more rapidly, we see improved productivity savings in our early results. Now Assist is especially significant because it gives us the capability to build things without the manual, repetitive work that is every employee’s nightmare.

“ServiceNow’s digital technologies operations team is at the forefront of using generative AI on workflows running on the Now Platform,” says Sankha N., senior vice president of digital technology operations at ServiceNow.

“We have already applied generative AI to several use cases and provide ongoing feedback to our product teams on the features. The more value we see this technology delivering on the platform, the more excited we are about future opportunities,” he adds. “As customer zero, we look forward to sharing these insights with our customers so we can learn and progress together.”

We’re especially thrilled to see the impact of generative AI on knowledge work through summarization, content generation, and code generation. AI on the Now Platform is vital to delivering value to our operations, as well as helping us orchestrate the entire technology value chain for our business.

Lessons from our generative AI adoption

Despite the hype around any new technology, we always return to our No. 1 priority: the security and privacy of our data. We need LLMs we can trust. Having LLMs integrated into the Now Platform removes a big obstacle. We have peace of mind that security and privacy are covered, so we can focus on improving other tasks, such as workflows.

Generative AI has also affected our staffing requirements. We don’t need as many machine learning experts to create AI models. Instead, we’re looking for good prompt engineers who can tweak generative AI models to suit various scenarios. As a result, we’re reskilling our team and changing our hiring criteria to embrace these new work demands.

There’s no doubt that generative AI is transforming ServiceNow, even in these early stages. We have a long list of internal use cases for this technology, and the list keeps growing. With the new capabilities in the Now Platform Vancouver release, we’ll continue to expand how we use the platform to increase efficiency, productivity, and value. Our quest is to bring intelligence to every corner of our business.

Find out how ServiceNow can help you put AI to work in your organization.