What I like about ServiceNow AI Agents is that they can automate repetitive tasks, provide quick responses, and help people solve problems faster. Review collected by and hosted on G2.com.

Sometimes, AI Agents don’t fully understand a user’s query or deliver completely accurate responses. For more complex issues, they may still require human support, so it’s important to have someone review the results when necessary. Review collected by and hosted on G2.com.

What I liked most was how quickly I could create tickets. The process felt very efficient: I could describe the issue in natural language, and the agent would generate a structured ticket fairly quickly. Review collected by and hosted on G2.com.

The agent sometimes miscategorized tickets or missed important technical context on more detailed SEO issues (crawlability, JS rendering, etc.). It also tended to default to a generic IT categorization rather than reflecting the specific SEO details. Review collected by and hosted on G2.com.

It works as a digital assistant within the ServiceNow platform and can take action automatically on its own to resolve issues. It also understands user intent, so it can solve problems instantly instead of relying on buggy scripts. Review collected by and hosted on G2.com.

Doesn't fit well with other tool instead of ServiceNow and human governance is always required while running multiple agents. Review collected by and hosted on G2.com.

Implementing ServiceNow AI Agents has been a game-changer for our IT Service Management. I love how the GenAI-powered agents not only provide links to articles but also summarize resolutions and guide users through workflows in plain English. We’ve seen a 30% reduction in Tier 1 tickets because the agents handle routine queries instantly, which has significantly improved our employee experience with 24/7 human-like support. The most compelling aspect is their evolution from simple assistants to autonomous "doers," handling complex workflows across the enterprise. I also find the Contextual Memory feature exceptional; it transforms the AI from a basic bot into a digital colleague that understands our specific environment. Review collected by and hosted on G2.com.

While ServiceNow AI Agents are powerful, they aren’t perfect. It’s only as smart as your notes: If your company’s help guides are old or messy, the AI will give bad advice. It doesn't 'know' things on its own; it relies on your data being perfect. Hard to set up: You can't just 'turn it on.' It takes a lot of time and technical experts to build the workflows and make sure the AI knows what to do. Review collected by and hosted on G2.com.

Help me search for dedicated teams link for reporting incident. For example, payroll, application access or escalating a request to dedicated team. It saves a lot time. Review collected by and hosted on G2.com.

Sometime the agent doesn’t provide accurate results. Review collected by and hosted on G2.com.

For common, well-known issues, an AI agent can be set up to provide the correct answer quickly. Review collected by and hosted on G2.com.

When a human talks to an AI, the amount of context they provide to the agent drops dramatically compared to when they speak to another human. Because of that, you need to be very deliberate about guiding the AI on how to gather the information it needs; otherwise, it’s easy for an almost empty ticket to get logged. Review collected by and hosted on G2.com.

I like that ServiceNow AI Agents act as true digital coworkers that don't just answer questions but autonomously execute complex, multi-step workflows across disjointed corporate systems to resolve issues end to end. It makes my life easier by actually doing the work for me, like fixing a password or ordering a laptop instantly instead of just giving me a link to read. Review collected by and hosted on G2.com.

It performs poorly if your underlying knowledge base or system data is disorganized and pricing can be unpredictable. Also, the long and tedious setup and struggle with complex requests. Review collected by and hosted on G2.com.

I like ServiceNow AI Agents for its efficient and pretty good responsiveness time, which helps me automate repetitive tasks. I appreciate the features like intelligent routing, suggested responses to manual work, and ticket triaging, which assist in providing support to our customers. These features help me reduce the need for repetitive tasks, speed up response times, and maintain consistent workflows. Review collected by and hosted on G2.com.

I found that the setup and customization takes some time and has a learning curve when first getting started. It wasn't the easiest, I would say it took some time to do. If you could have out-of-the-box templates for us to use, that would be good in my opinion, or a generally simpler workflow setup. Review collected by and hosted on G2.com.

I appreciate how ServiceNow AI Agents helps in automating workflows, which in turn optimizes costs and facilitates a faster turnaround time. I find it easy to configure and use, which makes the whole process smoother. I also like the summarization and resolution note generation features, as well as the change risk explanation, which are used repetitively by employees and add value to each ticket. The initial setup was relatively easy, which is always a plus. Review collected by and hosted on G2.com.

I don't like that the recommended action doesn't allow me to search by product and area in Yokohama. Review collected by and hosted on G2.com.

I use ServiceNow AI Agents for my work workflow, and it's awesome to use. Its tips are easy to follow. The biggest value is the reduction in manual effort and noise, especially in a data platform like Cloudera. When multiple alerts come from the same root issue, the AI agent intelligently deduplicates them into a single incident, which prevents alert fatigue. It ensures the right team is notified instantly through emails or integrations, improving response time. Review collected by and hosted on G2.com.

One area where ServiceNow AI Agents can be improved is accuracy in early stages. The predictive models depend heavily on historical data, so when the system is newly set up or when new types of incidents occur (like new Spark pipeline failures in Cloudera), the AI may misclassify incidents or assign incorrect priorities. It gets better over time, but the initial tuning phase can require manual intervention. Review collected by and hosted on G2.com.