Blog
10 steps to get started with AI in the workplace
AI is happening right here, right now. But when things are moving so fast, it’s easy to feel like you should be doing everything—preferably yesterday. To get started with AI in the workplace, you need above all to identify the right problems, create clear guidelines, get a handle on your data, involve your employees, and start testing on a small scale. After all, AI is a tool that can contribute to the solution—not the solution to everything.
Here are ten steps to help you get started. You can also take our self-assessment to see how AI-ready your organization is.
1. Start by identifyingthe problem
The first step is to identify a problem where AI External link. can actually be useful.
It’s easy to start at the wrong end: we have to do something with AI. But what?
Flip the question around. Where are the problems, the time wasters, and the frustration today? Maybe employees are spending an unnecessary amount of time searching for information. Maybe editors are doing the same manual task over and over again. Or maybe customer service gets the same questions every day.
Start with a real need, and then explore whether AI can help you solve it. That increases the chance that the technology will actually be useful.
2. Identify tasks where AI can provide support
AI doesn’t have to take over an entire task to create value. Often, the technology can be most useful by supporting people in specific parts of their work.
This could involve finding information more quickly, summarizing long texts, analyzing statistics, creating a first draft, or guiding a coworker through a process.
Together, these small improvements can free up a lot of time for the kinds of tasks that people are actually better at.
3. Establish clear guidelines for AI
To use AI safely, the organization needs clear guidelines on how the technology may be used. An AI policy or equivalent guidelines should describe which AI services may be used, what they may be used for, and what information may be shared. It must also be clear what information must absolutely not be fed into external AI services.
Guidelines shouldn’t just set limits; they need to help employees do the right thing. Because if the approved path is too complicated, there’s a risk that people will find an easier way on their own.
4. Get a handle on data and information
For AI to deliver good results, it needs to work with the right information. The old adage “garbage in, garbage out” is more relevant than ever.
Review what data and information you have, where it’s located, and how it can be used. Consider what happens to the data sent to various AI services. Where is it stored? How is it processed? Who has access to it?
At the same time, the content itself needs some attention. Old documents, duplicate versions, and conflicting information don’t become any less problematic just because AI can find them faster. On the contrary.
5. Make your content understandable to both people and AI
Good structure helps people, search engines, and AI services find and understand your content.
It’s no longer just people and traditional search engines that need to be able to interpret what you publish. AI services are increasingly acting as an intermediary between your information and those searching for it. Therefore, use clear headings, understandable text, relevant metadata, and a well-thought-out structure. Important information should be easy to find, understand, and use, even when taken out of its original context.
And don’t forget to clean up. AI is good at many things, but it shouldn’t have to act as a digital archaeologist.
6. Start with a small pilot project
When you start testing AI, you don’t need to choose the largest and most complex process in the organization.
Choose a clear and well-defined problem where you can test, measure, and learn. This could be, for example, an AI assistant that helps employees find information in the employee handbook, AI support for editors, or help with analyzing web statistics.
A smaller pilot project makes it easier to understand both the benefits and the challenges before you scale up.
7. Train and engage employees
Employees need both knowledge about AI and the opportunity to help shape how the technology is used. Give them the chance to learn how AI works and how it can help them in their own work. Talk to them. Organize workshops. Create pilot groups.
Also, study how people actually work—not just how they say they work. Often, the best AI ideas are found right in the midst of everyday life: in an unnecessarily complicated workflow, a recurring question, or something everyone complains about during coffee breaks.
8. Measure and evaluate your AI initiatives
Decide right from the start what you want an AI initiative to improve. This will make it easier to determine whether it’s actually creating value. Should employees save time? Should more people be able to find the right information? Should editors be able to publish faster? Or should the quality of the content improve?
Test on a small scale and track the results. What worked? What didn’t work? What needs to be adjusted? A pilot project doesn’t have to be perfect. The point is to learn. Once you know that something creates value, you can take the next step with much greater confidence.
9. Appoint people in charge and create a plan for the future
Sustainable AI work requires someone to keep everything together. Decide who will drive the work forward, gather insights, and ensure that guidelines, knowledge, and initiatives are aligned.
Also, develop a plan for how you want to expand the use of AI. The key here is to know where you want to go and why.
10. Maintain your human tone and identity
AI can help you create content faster, but you’re still the ones who need to provide the knowledge, perspective, and personality.
When AI can produce text, images, and other content in a matter of seconds, it becomes easy to produce more. But more isn’t automatically better.
In a world filled with AI-generated content, the human touch may actually become even more important. Your knowledge, tone, experience, and identity are what make you who you are. So use AI to enhance what you already do well, not to erode what makes you unique.
Start somewhere, and start now
Becoming AI-ready isn’t a project with a clear end date. Technology continues to evolve, and we continue to learn as we go. So don’t try to do everything at once. Start with a specific problem. Create a safe environment. Test something small. Measure the results. Learn from them and build on that.
Take our quiz: Is your workplace AI-ready?
You don’t need to have everything in place to get started with AI, but it’s good to know where you stand. Do you have clear guidelines? Are you keeping track of your data? Do you have the right skills and responsibilities in place? Take our self-assessment and get a quick overview of how AI-ready your organization is and what your next steps might be.
You Ask, We Answer
How do you get started with AI in the workplace?
Start by identifying a specific problem or task that could be improved. Establish clear guidelines for how AI may be used, review your data, and then test the technology on a small scale. Evaluate the results before moving forward.
What should AI guidelines include?
Among other things, the guidelines should describe which AI services employees are permitted to use, what they may be used for, and what information may be shared. They should also make it easy for employees to understand how to use AI safely and responsibly.
How can AI be used in the workplace?
For example, AI can help find and summarize information, create first drafts, analyze data, answer frequently asked questions, and guide employees through various processes. Start with the organization’s needs rather than the technology.
How can you tell if an organization is AI-ready?
An AI-ready organization has, among other things, knowledge of how AI can be used, clear guidelines, control over its data, and a process for testing and evaluating AI initiatives. It is also important to have someone accountable for AI-related matters and a plan for how the work will be carried out.