AI in Communications, Part 2: Implementing AI Without a Strategy

In my first post in this series, I wrote about how “AI” has become a vague catch-all term in many business conversations. That lack of clarity matters. If we are not specific about the problem we are trying to solve, it becomes very easy to chase tools before we have a plan.

And lately, this is something I hear frequently from employees when they talk about their company’s AI adoption.

Buying AI tools is starting to feel a lot like buying a corporate intranet used to feel.

There is excitement. There are big promises. There is often a polished demo. And there is a belief that once the tool is in place, better work will somehow follow.

Usually, it is not that simple.

Too often, organizations are starting with the tool instead of the goal.

That is backwards.

Strategy should come before the tool

A useful AI strategy does not begin with, “Which platform should we buy?”

It starts with questions like:

• What problem are we trying to solve?

• Where are people losing time today?

• What goal are we trying to achieve?

• What risks do we need to account for?

That investigation matters because not all AI tools are built for the same purpose.

Some tools are designed to support process and workflow. Others focus on content creation and ideation. Some are better at summarizing information or searching knowledge.

If organizations skip the strategy conversation, they often end up with tools that look impressive in a demo but do not actually fit the work.

The common mistakes I see

When organizations move too quickly, a few patterns tend to show up.

Tool-first thinking

The conversation starts with the platform instead of the problem.

It usually sounds like:

• “We should get AI.”

• “We need Copilot.”

• “Our competitors are using this.”

None of those statements define a strategy.

Vendor hype

AI vendors are selling possibility. That is their job.

The challenge for organizations is separating a compelling demo from a real business case. A strong presentation can make almost any tool look transformative, but the real questions are more practical:

• Will employees actually use it?

• Does it fit our workflows?

• How will we measure success?

No behavioral integration plan

Even a good tool will struggle if there is no plan for how people should use it.

Organizations need to define:

• Who is expected to use it

• What types of work it supports

• What review process exists

• What training employees will receive

Technology adoption is rarely a systems problem. Sometimes the challenge is access or usability, but just as often it comes down to whether people understand how the tool fits into their work and what they are expected to do with it.

Strategy also means understanding impact

Another piece that often gets overlooked is impact.

Before adopting AI tools, organizations should think carefully about what those tools will have

access to and how employees will use them.

For example:

• Do your privacy policies allow employees to enter company information into open AI tools?

• What internal data could be exposed unintentionally?

• Which tools are approved and which are not?

It is also important to establish clear ground rules early.

Employees need guidance on:

• What information can be shared with AI tools

• What outputs require human review

• How AI should and should not be used in their work

Without that guidance, teams can unintentionally misuse tools or generate large amounts of low-quality content.

A thoughtful strategy helps prevent that.

Explore, but do not overcommit

This is also a moment for balance.

Organizations should absolutely use this time to explore AI tools. The technology is evolving quickly, and there is real value in experimentation.

But exploration does not mean committing too quickly to one solution.

A smarter approach is to:

• Start with small pilots

• Test a few use cases

• Learn what actually improves work

• Expand only when the value is clear

The tools will continue to change. A clear strategy allows organizations to adapt without constantly starting over.

Where communications can help

This is one reason communications teams should be involved in AI discussions early, not just after a tool has already been selected.

Communicators are often uniquely positioned to help organizations move from broad enthusiasm to practical implementation.

They can help:

Define the strategy

• What problems are we trying to solve?

• What outcomes are we trying to achieve?

• How will success actually be measured?

Translate the strategy

• Explain why the organization is adopting AI

• Clarify how it will support work rather than replace it

• Set realistic expectations about what the tools can and cannot do

Create the structure around adoption

• Governance and guardrails

• Roles and responsibilities

• Clear expectations for employees

• Training and behavior change

Communications teams can also help measure adoption and impact:

• Are employees actually using the tools?

• Are they improving efficiency or quality?

• Are teams using them in the ways the organization intended?

There is also an important human dimension.

When AI is introduced without thoughtful messaging, employees often assume the worst. They worry about job security, replacement, or being expected to produce more with fewer resources.

Clear communication can help frame AI more accurately:

• As a tool that supports work

• As a way to remove repetitive tasks

• As a way to help teams focus on higher-value thinking

When strategy, messaging, and adoption are aligned, AI initiatives have a much better chance of delivering real value.

Final thought

Organizations should absolutely explore AI. But tools alone do not create value.

Before rolling out platforms, organizations need structured conversations about the problems they want to solve, the risks they need to manage, and the behaviors they want to encourage.

Otherwise, they are not really implementing a strategy.

They are just buying software.

The technology will keep evolving. Clear thinking about how to use it should evolve just as quickly.

If your organization is starting to explore AI and wants help defining the strategy, communicating the approach to employees, or measuring adoption and impact, I’m always happy to have that conversation.

And if there is a topic you would like to see covered in a future post in this series, feel free to message me. I’d love to include the questions others are wrestling with as well.