ServiceNow in practice: AI and the challenges faced by the people making technology work
A conversation about development, AI, security and the people who make it all happen.
☕ It’s the weekend. Enjoy the read with your favourite drink. We have some good stories from the ServiceNow Community to talk about.
Imagine starting the week with a simple idea: building an application to solve an everyday problem.
You open your personal development environment, look for App Engine Studio and expect to get to work.
Instead, you find a message saying that you need a valid licence.
Hang on — wasn’t I here precisely to learn? 😅
Meanwhile, in another community conversation, professionals are discovering AI assistants that can carry out tasks across systems, consult project information and help handle support tickets.
Two very different situations.
One is about the difficulties encountered along the way. The other presents new possibilities.
And both help answer an important question:
How do we turn technology into reliable outcomes for people and the business?
That was the thread running through our reading of the ServiceNow Community for the week of 18–24 October 2026.
Let’s get to the stories.
1. App Engine Studio: when learning meets an obstacle
Where did this conversation come from?
Source: user reports in the ServiceNow Community, rather than an official announcement of a widespread platform issue.
What the post presents: the author describes a licence message when opening App Engine Studio on the Australia release. They say they have the admin role and tried repairing the application without resolving the problem. Other participants share experiences and suggestions.
The story behind the message
Let’s imagine a professional learning to develop ServiceNow applications.
They have set aside a few hours to practise, opened their Personal Developer Instance (PDI) and tried to access App Engine Studio.
The aim was to learn how to build an application.
But a licence message appeared.
A PDI is a personal environment intended for learning and experimentation. App Engine Studio is a ServiceNow tool for building applications.
In the original report, the user explains that they can access ServiceNow Studio but continue to encounter difficulties with App Engine Studio.
That distinction matters because it shows that this is not necessarily a complete loss of access to the environment.
Other people joined the conversation, but there is not enough evidence to say that every case has the same cause.
And here is a valuable lesson.
A technical problem can look small yet have a big impact on someone who depends on that tool.
Imagine someone trying to develop a professional skill and discovering that part of their study time has to be spent investigating the environment itself.
It is frustrating.
But it is also an opportunity for the community to share knowledge and help clarify situations that documentation does not always resolve immediately.
What can we learn?
It is important to separate the observed problem from its possible explanations.
The report confirms a difficulty accessing that environment. It does not establish a widespread platform fault, a licensing change or a universal fix.
So our guidance is straightforward:
🧭 NORA Guidance | Development When you encounter a similar problem, record the release, the message received, the behaviour observed and the steps already tried. Compare that information with the documentation and existing reports before making changes. A good investigation starts with understanding the problem, rather than the first solution you find.
Learning begins before the first line of code.
It begins when someone can access the tools they need to create.
2. Artificial intelligence: when the conversation moves beyond the chatbot and into work
Now let’s change the scene.
Imagine someone arriving at the office to start the day.
They need to prepare for a meeting, check important tickets, consult project information and reply to messages.
The data is available, but spread across different applications.
The challenge is not simply finding information.
It is bringing the context together, understanding what needs attention and taking action.
During the week, posts by ServiceNow employees presented AI capabilities related to those needs.
Let’s look at three examples.
ServiceNow Arc: working across applications
Where did this news come from?
Source: an article by Paty Montesinos, identified in the ServiceNow Community as a company employee.
What the article presents: ServiceNow Arc is described as a personal AI assistant that coordinates activities across different applications. The post includes examples of meeting preparation, operational reports, handovers between teams and governance controls.
And how does that work in real life?
Think about preparing for a weekly meeting.
You need to consult appointments, messages, documents and ServiceNow records.
That preparation usually involves several lookups and some manual organisation.
Arc is presented as a way to coordinate that work across systems, using the professional’s authorised access.
But there is an essential detail.
The assistant should not have unrestricted freedom to act simply because someone asked.
According to ServiceNow, Arc uses controls to check permissions, approvals, execution limits and access to capabilities.
That is particularly relevant when AI can take action rather than just produce answers.
The post also sets out specific compatibility, licensing and usage conditions. The general availability announced in October does not remove those requirements.
🎬 Watch the demonstration
What to expect: the post presents the video as a demonstration of the Arc enterprise assistant. The article explains its role in coordinating tasks and applying controls.
The demonstration is supplementary: it helps visualise the proposal we have just discussed.
💡 What really matters here?
It is not just about having an assistant that can open applications and carry out activities.
It is about reducing operational effort without losing control of information and decisions.
🧭 NORA Guidance | Enterprise AI Before considering an assistant such as Arc, choose a recurring activity that takes up time. Identify the systems involved, the access needed and the decisions that require human participation. Start with the work that needs to improve. Then assess the technology.
MCP for SPM: when a project needs to tell its own story
Where did this news come from?
Source: an article by Vinitha Chukkala, identified as a ServiceNow employee.
What the article presents: MCP for SPM connects goal, portfolio and project information from ServiceNow Strategic Portfolio Management to compatible AI assistants. The post describes the tools available, access controls and requirements for use.
The question every manager knows
Imagine a project progress meeting.
Someone asks:
Why is our project behind schedule?
Answering responsibly may require looking at deadlines, risks, dependencies, activities and business goals.
Strategic Portfolio Management (SPM) helps organisations manage this kind of information.
Meanwhile, MCP for SPM allows compatible AI assistants to consult authorised SPM information through the Model Context Protocol.
It is important to distinguish the concepts:
MCP is an open protocol, used to connect AI tools to systems and data.
MCP for SPM is a specific implementation offered by ServiceNow.
In the post, the company describes capabilities for consulting projects and goals, analysing risks and preparing progress information.
Access respects the authenticated user’s permissions.
That means connecting an assistant should not allow someone to see data they would not normally be able to access.
🎬 Watch the demonstration
What the demonstration shows, according to the post: Claude with MCP for SPM in three situations: a manager investigating delays, an owner tracking business goals and a portfolio manager identifying projects at risk.
This is a good example of a video that adds context to the reading.
It connects a technical capability with the questions professionals ask in everyday work.
But we must not forget one point.
An AI-generated answer still depends on the quality of the information used.
If records are incomplete or out of date, the answer may look excellent without necessarily reflecting the real situation.
🧭 NORA Guidance | Portfolio management Before connecting an assistant to project data, assess the quality of the information, who is responsible for updating it and which decisions it will support. AI can make knowledge easier to access. Responsibility for keeping that knowledge reliable remains human.
L1 IT Service Desk AI Specialist: what happens after a ticket arrives?
Where did this news come from?
Source: an FAQ by Christopher Ell, identified as a ServiceNow employee, with a revision recorded on 20 October 2026.
What the content presents: explanations of the AI Specialist’s role, how it differs from chatbots, support activities, requirements and availability conditions.
Let’s think about a support team
An employee cannot access an application.
They raise a ticket.
Someone needs to interpret the request, investigate the situation, consult information, try to resolve the problem and document the result.
In many cases, these are familiar, repetitive activities.
The L1 IT Service Desk AI Specialist is presented by ServiceNow as a capability for carrying out first-line support activities.
The proposal goes beyond answering questions.
According to the FAQ, the capability can receive assigned tickets, diagnose situations, perform permitted actions, document the support provided and escalate cases when necessary.
That differs from the traditional role of a chatbot, which usually focuses on conversational interaction.
But we need to make an important distinction.
Automating activities does not remove the need for people.
A support team also handles exceptional situations, expectations, communication, priorities and decisions that require judgement.
The capability has specific usage conditions and initial availability for selected customers.
🎬 Explore the proposal
The FAQ includes a playlist of videos about the L1 IT Service Desk AI Specialist.
We do not need to turn every article into a video gallery. Audiovisual material should help readers understand something that the text alone cannot explain as easily.
And here is one of this week’s most important reflections.
What kind of work do we want to automate? And what kind of experience do we want to preserve?
🧭 NORA Guidance | IT support Start by identifying repetitive tickets, the information needed to resolve them and the situations that require human intervention. Then assess whether automation can improve the experience and outcomes. The aim is not simply to reduce the number of human tasks, but to let people devote attention to the work where they contribute most.
3. Security: when identifying a vulnerability is only the beginning
Let’s imagine a security team receiving an advisory about critical vulnerabilities.
The first challenge is understanding the situation.
Then finding out which systems may be affected.
Finally, organising the response.
During the week, a post discussed what is known as PHP2Shell, relating to vulnerabilities in PHP.
Where did this news come from?
Source: an analysis published in the Armis area of the ServiceNow Community, with authorship attributed in the text to Michael Freeman and Guy Broder, Armis specialists.
What the article presents: an explanation of the vulnerabilities, their potential impact, ways to identify exposed environments and mitigation measures.
The text is hosted in the community, but that does not make PHP2Shell a vulnerability in the ServiceNow platform.
PHP is a technology external to ServiceNow.
The vulnerabilities were disclosed by the PHP project, which also published security updates.
What does this story teach us?
An organisation can receive a technically accurate alert and still struggle to respond.
Where are the affected systems?
Who is responsible for them?
Which ones are exposed?
Which fixes need to be prioritised?
These questions are part of security work.
And it is in this context that asset management processes and solutions such as ServiceNow Vulnerability Response can support the organisation of vulnerability remediation.
Vulnerability Response does not replace the fixes supplied by the PHP project or the vendors of affected systems.
It can help organise information, responsibilities and follow-up, depending on the capabilities available and configured.
The outcome depends on both the technology and the people carrying out the process.
🧭 NORA Guidance | Security Use official PHP advisories to confirm the affected environments and applicable fixes. Identify exposure, involve the owners and track the measures through to completion. Security requires attention and coordination. We do not need to turn risks into a spectacle, but we should not ignore them either.
4. What do these stories really teach us?
Perhaps you are wondering:
What do a development problem, three AI capabilities and a PHP vulnerability have in common?
Technically, they do not have to share anything.
They are different situations, with distinct origins, consequences and levels of evidence.
It would not be right to call them a single market trend.
But we can learn from all of them.
In the development experience, someone needs to be able to access a tool to learn.
In project management, someone needs reliable information to make decisions.
In IT support, someone needs appropriate help to resolve a problem.
In security, someone needs to identify risks, involve the owners and track remediation.
In each of these stories, there is a person trying to accomplish something important.
And technology plays a part in that journey.
That is why I believe in three principles for assessing a platform’s evolution:
Clarity, to understand the problems, possibilities and limits of solutions.
Security, to protect information, respect permissions and reduce risks.
Responsibility, to know who decides, who acts and how outcomes are followed up.
These are not just technical concerns.
They are commitments to the people who depend on systems every day.
Before we finish our conversation...
We do not need to start with the newest capability.
Nor with the most talked-about problem.
Let’s start with the work that needs to be done.
With the people involved.
With the information needed.
With the outcome that will really make a difference.
Then we assess the role of technology.
Because, in the end, we are not just configuring systems.
We are helping people work better, overcome difficulties and deliver outcomes for other people. 🤝
And that is why it is worth continuing to listen to the community.
See you next week. And enjoy the rest of your weekend!
Sources and editorial transparency
User reports — App Engine Studio
Posts by ServiceNow employees
Security — Technology external to the ServiceNow platform
Editorial note: this article interprets discussions and posts relating to the week of 18–24 October 2026. Participants’ reports do not represent official ServiceNow statements. Product descriptions reflect the publications and conditions stated by the company. Everyday scenarios are illustrative examples, not verified personal experiences. The research coverage does not represent the entire community or market.
Content consolidation and writing support: NORA — Market Signals.
Editorial review and approval: André Costa Moreira.
