ServiceNow generative AI enhances productivity, streamlines workflows, and creates exceptional customer experiences

According to recent research conducted by McKinsey, the implementation of generative AI in 63 analysed use cases could result in an annual economic value ranging from $2.6 trillion to $4.4 trillion.

In customer service, generative AI can transform the customer experience by enabling personalised interactions, automating routine tasks, and providing real-time assistance to human agents, ultimately leading to improved customer satisfaction

In following, let’s review the capabilities of ServiceNow Now Assist, how it works, and how it supports service management and operations.

ServiceNow Now Assist Capabilities

Now Assist as a set of tools and integrations that can be used at various places of a ServiceNow platform. Specifically, you can interact with ServiceNow LLM or 3rd party LLMs and integrate them into the service management, operational, support, and app development processes. In other words: can generate new content from text inputs like search boxes, text fields, or chat.

As of now, there are following capabilities available:

ServiceNow offers pre-configured Now Assist extensions that you can install from ServiceNow Store. These extensions provide the actual capabilities for various use cases. Currently, there are assists available for IT Service Management (ITSM), Human Resources (HR), Field Service Management (FSM) and Customer Service Management (CSM).

These apps connect with the Generative AI Controller, a low-code integration that links Generative AI Application Programming Interfaces (APIs) to the Now Platform. With the Generative AI Controller, the Now Assist applications can communicate with large language models (LLMs). These apps allow users to use generative AI features (called Skills) specific to ServiceNow.

What problems are solved with ServiceNow generative AI?

  • Knowledge article curation
  • Search result accuracy
  • Summarization of complex information
  • Detection of customer and employee sentiment
  • Accuracy and scalability of custom content
  • User experience and satisfaction
  • Code generation
  • Agent handoff 
  • Errors and misunderstandings between users and agents

How to prepare your Data to have most value from Now Assist?

Similar to all other services, like OpenAI, ServiceNow LLM requires data to create, train, and continuously improve the models used for automation and prediction. The accuracy of the models is dependent on good-quality data. The data needed is dependent on the capability being implemented.

If you use Generative Search, you should take care of high quality Knowledge Base, means, it’s content has contain correct information about problems and known solutions applicable for your organisation. Also, ensure you continuously improve and update the content.

For summarising Incidents and Cases, take care of Knowledge articles, Ticket descriptions, notes, work notes, and comments Interactions (e.g., chat).

Now Assist panel relies on information within Ticket descriptions, notes, work notes, and comments Interactions (e.g., chat)

When developing business rules and script includes, the Code is generated based on the comments in a script.

Data Guidelines

Where is ServiceNow Now Assist available?

ServiceNow Now Assist is not available per default. You need a subscription in order to be able to install and run Now Assists for ITSM, HRSD, CSM, FSM etc. ServiceNow offers a new subscription model ServiceNow SKU Pro Plus. If you are a ServiceNow partner and not using ServiceNow productively, you should contact your partner manager.

In general, ServiceNow Now Assist is available on Customer Instances and Self-Hosted Instances. It is not available, on Personal Developer Instances (PDI), GCC, and FedRamp environments.

Which generative AI Providers does ServiceNow support?

ServiceNow has its own LLM and also supports OpenAI and Microsoft Azure OpenAI. That means, you can include non-ServiceNow AI capabilities into your workflows. You can use following third-party AI providers configuring Generative AI Controller:

Within the Generative AI Controller you can configure APIm endpoints, adjust model properties, include into ServiceNow Virtual Agent and other workflows. There is, of course, a risk sharing sensitive data with a 3rd party AI service. However, ServiceNow can handle the personally identifiable information and other sensitive data and exclude from appearing in generative AI prompts. Instead, placeholders are sent and then replaced with the original text after the response has been received. This two-way masking ensures your users see the correct values, but the LLM is not exposed to any sensitive information.

The integration is based on Integration Hub. However, the transaction will not be included into the Integration Hub transaction costs as it consumes “Assists” as part of the Pro Plus/Enterprise Plus SKU.

Which releases are compatible with the Now Assist?

Customer instances should be on Vancouver Patch 2 or higher. However, in order to start using Now Assist and its Generative AI capabilities you will need to upgrade to ServiceNow Professional Plus or Enterprise Plus SKU. You should contact your Account Executive for further details. If you need any assistance, feel free to reach out to us.

Which builder interfaces use generative AI?

You can use generative AI within Virtual Agent Designer, Flow Designer, and Scripting.

Beside that, there are built-in capabilities like summarisation that can be used ad-hoc within incidents, cases etc. AI Search also uses generative AI to better understand users request and provide better search results.

How does ServiceNow handle data privacy?

When using ServiceNow LLM, the data stays in the ServiceNow environment. When using the Generative AI Controller, your data and queries will be sent to OpenAI and Azure OpenAI. ServiceNow recommends you peek at their data privacy policy and decide what usage policy works best for your organisation. Further details under OpenAI Usage Policy and Data, privacy, and security for Azure OpenAI Service.

What models are currently supported?

Currently, ServiceNow supports GPT-3, GPT 3.5 turbo, and GPT-4 models.

Is Generative AI relevant for your organisation?

Most likely, yes. Generative AI can be applied in any organisation and process. You have the most benefits with high volume and repetitive work. Beside that, any process where you experience a gap with people and talents, are worth to be automated as much as possible and let people work on more valuable tasks.

Check out our Webcast “How corporates leverage GenAI for better service management and operations?” to learn more about use cases and relevance of AI for corporates and enterprises.

Do you need support with ServiceNow AI or ServiceNow SKU Pro Plus?

Our professional can support you in understanding and evaluating ServiceNow capabilities. We built various solutions and applications that include Generative AI and know how to apply it with the highest outcomes. Check out our AI offering for further details.


Teiva Systems is a ServiceNow Premier Partner and provides advisory, architecture, implementation, training and support services. Teiva Systems specialises in building custom applications and integrations with ServiceNow App Engine leveraging platform capabilities like common data model, security, integration layer, now assist etc.

Kostya Bazanov is ServiceNow Leading Architect and Technology Advisor since 2011. He acted in various roles such as technical consultant, implementer, solution architect, team lead in national and global projects and ServiceNow platform rollouts. Kostya is certified in ServiceNow service management, operations management, custom applications.

Kostya Bazanov, Managing Director, Feb 26, 2024

Eager to take the next step? Contact us today!

* Required fields

Latest Articles

ServiceNow Xanadu: Release Notes, New Applications, Platform Capabilities, Hidden Features

ServiceNow Xanadu: Release Notes, New Applications, Platform Capabilities, Hidden Features Shangdu, more popularly known as Xanadu, was the summer capital of the Yuan dynasty of China. Shangdu is located in the present-day Zhenglan Banner, Inner Mongolia. The city was a cultural melting pot that played host to visitors from throughout Asia and even further afield. […]

read more

What is the difference between DORA and MaRisk, BAIT, NIS 2 and Supply Chain Act?

In the previous article about DORA “When do financial and ICT companies have to be DORA compliant?” I took a look on what is DORA and what system capabilities do you need to enable and support the regulatory requirements of the Digital Operational Resilience Act. DORA establishes detailed rules for operational risk management. But how […]

read more

A better Compliance Management with GenAI and ServiceNow (Webinar)

Lately, we’ve been diving deep into a new use case we developed leveraging Generative AI (GenAI) within Compliance Management. Understanding the challenges of a industry and having an experience of solving complex problems is the key, when your goal is to help clients transforming their operations and business. Companies in Energy and Utility sector face […]

read more