What to know before forming an AI task force at your organization

From huge companies to local governments, it seems like every organization has an AI task force. 

When an org decides to set up an official AI group, it’s often because leadership or boards see the trends and wonder why their company isn’t keeping up. But the work doesn’t stop at just putting together a team.

Unless the whole organization is on the same page and goals are set out ahead of time, a task force won’t be successful, Thomas Elliott, founder of guidance group Augmented Intelligence Advisory Group, told Technical.ly. 

“The more questions you ask up front, the better informed you’re going to be as to how it is your company actually runs, and what are the different opportunities that we have to be able to add AI to accelerate that,” he said. 

Here are the questions to ask, first steps to take and how to measure success, according to experts.

➡️ Jump to a section:
• What is an AI task force?
• Why would an organization form an AI task force?
• What does that look like in practice?
• How can I start a task force?
• Who should be on an AI task force?
• How is a task force different from a working group?
• What kinds of goals should AI task forces be setting?
• How much authority does an AI task force have?
• What are the drawbacks to an AI task force?
• Is a task force the most effective method for bringing AI into a company?

What is an AI task force? 

An AI task force is an internal group focused on answering questions related to AI, Brian Green, chief AI ethics officer at AI governance startup Health-Vision.AI, said. 

The goal of the group is to identify the opportunities and challenges that would come with implementing AI at the company, according to Elliott from the AI advising group. Leaders of the task force set the company’s goals and assess how AI can be used to help achieve them.

In other words, they figure out how AI could improve efficiency while being implemented responsibly. 

They could do this by taking advantage of AI tools in the software the company already uses, or researching new technologies that could be beneficial. 

These groups largely focus on education and awareness, ensuring employees have access to the technology and know how to use it.

As AI takes off internally, the task force usually tracks impact. The group establishes how to measure if AI is actually saving the organization resources and if it’s helping employee workflow.  

Why would an organization form an AI task force? 

Organizational leaders form a task force when they’re looking to incorporate AI into the firm, but they still need to learn as much about it as possible. 

The task force is the first step in the discovery process to get to know what’s out there and convene people from across the business who want to help champion the technology. 

Then, they can turn it into actionable plans, implementing AI in different parts of the business and upskilling employees as needed, Andreas Welsch, founder of AI consulting company Intelligence Briefing, said.

For example, an organization may need a task force to help teach employees how to use AI for personal productivity, but it can also help with finding ways to make complicated processes — like getting insights from data or screening job candidates — more efficient.

What does that look like in practice?

The nonprofit Alliance for Decision Education established a task force to manage distribution and education around AI tools in 2024. 

AI presented an opportunity to do more with the org’s limited resources, Charles Cassidy, organizational development senior manager, said. 

First, the task force selected a platform, encouraging employees to test out different use cases, he said. 

Throughout the process, the task force offered guidance to the org’s different teams as they learned how to use AI. For example, putting together a “menu” of different ways to use the tools and curating training sessions based on what was most interesting to each department. 

“You need people to do the work, … and advocate for [AI’s] ongoing presence,” Cassidy said. “The task force has that role as well in … making sure it remains top of mind.” 

So far, the nonprofit is seeing results.

Over the last year, there’s been an increased enthusiasm for AI among the org’s employees, according to Cassidy. There’s also been an improvement in the quality of the org’s work by creating more robust project proposals and including more perspectives, he said. 

How can I start a task force?

Before launching an AI task force, leaders need to do an internal audit.

Decide what the company’s goals are for the next few years and if the right tools and people are in place to accomplish them, Welsch, an AI strategist, said. 

“Whoever is in charge of the AI task force needs to figure out what is our overall business strategy or business goal that we want to achieve in the next three [to] five years,” he said. “How can we use technology like AI to reach these goals more quickly [and] cheaper.” 

Green, an expert on AI ethics, recommends starting with a companywide workshop to identify what people already know about AI, what they’re afraid of and how they think it could be used. 

“You start with this workshop-based approach, level-set expectations, get everyone on the same page, and then you form the task force, as an outcome of this type of meeting,” he said. 

Once the members are selected, the next step is spreading the word. A task force holds workshops to raise awareness about AI in the workplace and gather information on what would be most helpful. These sessions can happen slowly, first with executives, then mid-level leaders and finally entry-level employees. 

Who should be on an AI task force?

It depends on your organization. 

AI task forces look different across organizations, and can vary in size depending on how big the company is. 

One person should be designated to lead the task force, and the more senior they are, the better, Welsch from Intelligence Briefing said. Understanding the business across departments and having some level of decision-making power will help move things along. 

Then, a committee should be created with representation from all different parts of the business, including technology, HR, finance, sales, marketing and legal. This diversity means they can more thoroughly talk about ethics, bias, data privacy, regulations and more, he said. 

“Everybody has different objectives,” Welsch said, “but the goal is to make sure everybody also understands why the company is now focusing on AI.”

How is a task force different from a working group?

Terms like working group and center of excellence are also used to describe internal groups focused on AI. These terms can vary depending on the size of the company and how many resources the group has available to it, AI consultant Welsch said. 

A governance committee is specifically focused on setting up the rules and ethics around AI usage, which a general task force may not do, Green of Health-Vision.AI said. 

What kinds of goals should AI task forces be setting? 

Goals set by the task force can either be about personal productivity or measuring specific business metrics for the company, consultant Welsch said. 

First, the org should look at how many employees have access to the AI tools. Then it can measure how many employees are actually using them and how often they are using them, Welsch said. 

Once the conversation is started, the task force can find out people’s experiences working with the tools to find room for improvement. 

On the business metrics side, the task force would set specific, quantifiable goals, often called key performance indicators. Those indicators measure how departments are using AI to streamline their processes. 

A task force could be deemed successful if the company can provide metrics that show there was time and money saved, Welsch said. 

How much authority does an AI task force have?

The authority of a task force depends on who is leading it, consultant Welsch said. If the group is led by a C-suite executive or even a chief AI officer, there is more likely to be more impact. 

Some task forces are led by leaders in the IT department, but those figures also tend to have less authority, he said. 

What are the drawbacks to an AI task force? 

Without clear leadership and goals, it’s easy for a task force to fizzle out, Green from Health-Vision.AI said. 

A lot of task forces fail because the initial excitement disperses, AI consultant Welsch added. If there isn’t clear communication, consistent participation or strategic direction around AI efforts, employees will go back to doing things the way they used to. 

In fact, many task forces do not achieve the goals they set. McKinsey released a report in September that found that while 80% of companies are using AI, the same percentage did not see significant benefits. 

Leaders can avoid this by setting clear goals and a clear chain of leadership, Green said. 

Is a task force the most effective method for bringing AI into a company? 

It’s the most effective method right now, Welsch, who advises businesses on AI strategy, said, because it’s been tested and proven through different technology cycles. But they won’t be around forever, or else they’d turn into business units. 

Task forces are a transitional body, he said. They usually start out as a small group of experts that share information with the wider company. 

Over time, usually about two to four years, department leaders take learnings back to their teams, and the centralized task force shrinks. Individual upskilling and learning happen in each department.

“From a centralized model,” Welsch said, “it moves more into a federated or into a decentralized model as time goes on and these organizations scale and mature.”