AI Isn’t the ‘Easy’ Button: How to Identify and Plan for AI’s Capabilities in Your Business

As much as we wish it were true, AI isn’t an “easy” button. Instead, planning for AI is more like an odd game of Candyland that requires you to chart a path through the “Fog of FUD (fear, uncertainty, and doubt), over the mountain of “Inflated Expectations,” and across the “Valley of Disillusionment” before you finally reach the Peak of Improved Performance. While no two organizations’ paths will be the same and there is no standard playbook (yet), there are some obstacles all organizations will encounter and best practices for navigating them.

Step 1: Plan.

Planning is key, regardless of where your organization is in its AI journey. At the outset, consider how your organization handles and adapts to change. How do you typically handle change management? Do you find champions and allow them to evangelize within the organization? Or do you make tools available and expect they will be adopted? The answers to these questions will guide you as you plan to incorporate AI into your organization.

Don’t forget to budget properly. Most GenAI tools are pricey, so it’s crucial to understand the licensing terms, overall costs, and integration measurements. It is hard to know exactly what GenAI will look like in a year or two but putting pieces in place now will enable your organization to be prepared for wherever the technology goes.

Finally, proper planning requires doing your due diligence. Conduct comprehensive assessments of your vendors. Know how much of your data they are accessing and what they are doing with it. Understanding how your technology and data are impacted by your relationships with others, including vendors, will help you move to the next step of securing that data.

Step 2: Secure.

The first step to secure your organization is creating an AI Acceptable Use Policy that outlines what tools and tasks are appropriate for AI use. You should also determine which clients want you to use AI and which don’t so you can decide how to communicate those desires. If you don’t have a master data management plan, now is a good time to create one. GenAI is exceedingly skilled at finding nuggets of data you didn’t know existed, and “security by obscurity” is no longer an option. You must know the full scope of your data and who has access to it before you can unleash GenAI to your organization and its employees.

Step 3: Educate.

Ensure that your employees understand what AI can – and can’t – do. Educate your teams on AI’s strengths, weaknesses, and potential improvements. Knowing when not to use the tool is as important as knowing when to use it. So, help your employees understand how to spot hallucinations and craft prompts that will get the most useful responses. For instance, employees should be trained on how to incorporate the six factors of a quality prompt (task, context, exemplars, persona, format, and tone) to maximize efficiency and efficacy. 

Step 4: Use.

Finally, understand who is using AI, how they are using it, and why they are using it. Identify the business problem you are seeking to solve with the tool and develop new workflows to take advantage of the increased efficiency. Meanwhile, it’s important to allow room for creativity because people will find new and inventive ways to leverage the technology.

While GenAI can feel overwhelming, don’t give up early. You will likely have successes and failures as you incorporate the technology into your organization, but AI is too important and consequential to give up on it entirely if you don’t get it right at the get-go. Be willing to say “not now” or “not yet” without saying “never.” Be prepared for continual evolution. Unexpected problems will arise, but with the right policies in place, strong data protection mechanisms, and proper education, you can navigate those challenges. Ultimately, AI is a tool, not a solution. And because it is a tool that requires us to rethink existing processes, make sure to think before you act.