Context​

 

A small internal communications team in a large, global technology company supported more than 5,000 employees across regions and functions. They handled a steady stream of high-stakes work: digital transformations, organizational changes, M&A, major tooling shifts, and crisis communications.​

 

The pace was relentless. Requests often arrived with same-day or same-hour deadlines, and many topics affected compensation, roles, or ways of working. Despite this scope, the team comprised just five professionals plus a manager, with no ability to backfill during layoffs and attrition.​

 

Leadership expectations were clear: deliver more, faster, with high quality — despite shrinking capacity. AI looked like a potential lever, but only if it could be adopted without compromising trust, accuracy, or craft.​

 

Starting Point: Curiosity, Skepticism, and Real Constraints​

 

The team’s starting point mirrored many communications organizations:

• Some team members were eager to experiment with AI.

• Others were skeptical, believing their work was too nuanced or sensitive.

• A few expressed concerns about privacy, quality, and misuse.

• Everyone was already stretched thin.​

 

These reactions were treated as signals, not resistance. Any adoption effort would need to respect professional judgment, acknowledge workload realities, and create psychological safety for experimentation. AI would not succeed as a mandate; it would need to become a shared practice.​

 

Approach: Guardrails Before Scale​

The company had implemented a private, internal large language model that allowed teams to work with confidential material as long as content stayed in the secure environment.​

 

Even with that foundation, clear team-level guardrails were established:

• AI could support drafting, but not final outputs.

• All content required human review, fact-checking, and approval.

• Public LLMs were prohibited for internal communications.

• Past examples were required to anchor tone, consistency, and voice.​

 

At the enterprise level, IT and Legal reinforced strict privacy boundaries, ensuring content entered into the system was not accessible — even internally — beyond the user. That clarity eased concerns and helped build trust.​

Early Experiments: Practical Use Cases, Not Abstractions​

 

Instead of broad, abstract experimentation, the team focused on immediate, day-to-day use cases:

• Drafting internal communications and FAQs.

• Turning raw inputs (emails, transcripts, notes) into structured messages.

• Brainstorming approaches for sensitive communications.

• Creating multiple versions of the same message for different audiences.

• Synthesizing large volumes of employee feedback to identify themes.​

 

A short, required prompt-training course established baseline skills. One insight stood out: examples mattered. Feeding the system prior communications dramatically improved tone and relevance, reinforcing the importance of human context.​

 

AI’s value became clear: not as a replacement for communicators, but as a way to eliminate the blank-page problem and reclaim time for judgment and strategy.​

Human Dynamics: Coaching, Trust, and Skill Acceleration​

 

As adoption progressed, the most meaningful shifts were human, not technical.

 

A junior team member — the newest and least experienced — embraced AI early. With coaching, she produced work that previously would have required much more senior involvement; her confidence and contribution grew quickly.​

 

At the other end of the spectrum, a senior communicator initially resisted, convinced his work was too specialized for AI to add value. Through coaching and experimentation, he discovered that his deep expertise actually produced the strongest results when paired with thoughtful prompting.​

 

A pattern emerged:

• Junior talent used AI to stretch into more complex work.

• Senior talent used AI to accelerate strategic thinking and refinement.​

 

Rather than flattening skill differences, AI amplified them, in productive ways.​

Outcomes: Capacity, Consistency, and Strategic Focus​

 

Over time, the team saw tangible benefits:

• First drafts that once took hours could be produced in minutes.

• Multiple message options could be tested quickly.

• Communications became more consistent across initiatives.

• Leaders received faster, higher-quality support.

• Junior employees took on more responsibility.

• Senior team members regained strategic bandwidth.​

 

AI’s ability to digest large volumes of qualitative input — including verbatim survey comments — and synthesize clear themes proved especially valuable. It helped the team close feedback loops more quickly and respond to employee sentiment with greater precision.​

 

Most importantly, the team sustained performance during a period of declining headcount, without burning out or sacrificing quality.​

What This Case Demonstrates​

 

For internal communications functions, this case shows a realistic pattern:

• Responsible AI adoption is a change and capability challenge, not a tools rollout.

• Guardrails enable trust, which enables experimentation.

• Coaching matters more than enthusiasm.

• AI can accelerate both junior growth and senior effectiveness.

• Human judgment remains the differentiator.​

 

AI did not replace this team. It unlocked capacity, consistency, and confidence, while preserving craft and trust.​

 

How I Help In-House Communications Teams Do This Work​

 

This case reflects a pattern I see across internal communications teams under pressure to “do more with less” while navigating sensitive, high-stakes work.​

 

In my consulting and advisory work with in‑house communications leaders, I help teams:

• Design AI guardrails and usage principles that protect trust, privacy, and craft.

• Identify practical, high‑value AI use cases aligned to real work and constraints.

• Build skills through coaching so both junior and senior communicators can use AI well.

• Create feedback and learning loops that keep the function credible, responsive, and sustainable.

 

If your team’s reality sounds similar, we can adapt this kind of approach to your context — so AI becomes a supportive partner to your communicators, not another source of pressure.

Return to case studies page →