AI at Work: A Comms Guide to Using It Well (Without Freaking Everyone Out)

Let’s get this out of the way early: AI is already in your workplace.

‍

Whether your organisation has officially rolled it out, banned it, quietly tolerated it, or is pretending it doesn’t exist (bold strategy), people are already using AI tools to write emails, summarise documents, draft presentations and occasionally ask existential questions at 11pm.

‍

So, the question for comms teams in 2026 isn’t “Should we use AI?”
It’s “How do we talk about it sensibly, safely and without triggering mass panic?”

‍

Because handled well, AI can be a productivity booster, a creativity helper and a serious time-saver. Handled badly, it becomes a compliance nightmare wrapped in an anxiety spiral. And guess who sits right in the middle of that? Yep. Us.

‍

A considered, sensible and realistic comms strategy around AI is pretty much an essential in 2026.

For some companies, a separate and bespoke campaign might even be needed.

‍

‍

First things first: AI isn’t the enemy. Silence is.

‍

One of the biggest mistakes organisations make with AI is saying nothing.

‍

When there’s no clear guidance, people fill the gaps themselves. They experiment quietly, copy/paste sensitive information into tools they shouldn’t, or assume leadership is either clueless or hiding something.

‍

None of those outcomes is ideal.

‍

Our role as comms professionals isn’t to be the AI police. It’s to help the organisation set clear expectations, reduce risk, and build confidence, without turning every message into a legal disclaimer with a pulse.

‍

‍

How comms teams should frame AI from the start.

‍

The tone we set early really matters. (As does our tone of voice in all important comms.) If AI is introduced as mysterious, risky or vaguely threatening, employees will respond with fear, resistance or covert usage that would impress the most secret of secret squirrels.

‍

A healthier framing is this:

AI is a tool.
It’s not a replacement for people.
It’s not magic.
Nor is it something everyone must use all the time.

‍

And crucially: how it can be used depends entirely on your organisation, your sector, your data sensitivity, and your regulatory environment. That nuance needs to be baked into your comms from day one.

‍

‍

Helping employees understand what is and isn’t appropriate use of AI.

‍

This is where comms teams earn their keep.

‍

Most people don’t want to misuse AI. They just don’t know where the lines are - especially when those lines differ between companies, roles and industries. Our job is to make those boundaries clear, practical and easy to remember.

‍

A good starting point is to separate AI use into three buckets: generally OK, proceed with caution, and abso-bloody-lutely not.

‍

‍

Generally OK (in many organisations)

These are low-risk, high-value uses. [Guru Disclaimer: always subject to your own policies]

‍

• Drafting or polishing internal emails (without sensitive content)
• Summarising long documents already approved for internal use
• Summarising & minuting meetings (without sensitive content)
• Brainstorming ideas, headlines, structures or outlines
• Turning notes into clearer first drafts
• Creating generic templates or frameworks

This is where AI shines as a co-pilot, not an author.

‍

‍

Proceed with caution (policy-dependent)

These uses often require clear rules, training, or approval.

‍

• Working with customer data
• Using AI to analyse internal performance data
• Generating external-facing content
• Supporting decision-making that affects people (e.g. hiring, performance)
• Using AI outputs without human review

This is where comms and legal need to be aligned, and where “check before you use” should be a normal message, not a scary one.

‍

‍

Absolutely not (in most organisations)

These are the red lines employees need to understand clearly.

‍

• Uploading confidential, personal or commercially sensitive data into public AI tools
• Using AI to bypass internal controls or approvals
• Presenting AI-generated content as fact without verification
• Using AI to impersonate colleagues or leaders
• Letting AI make final decisions that require human judgement

If these boundaries aren’t spelled out plainly, people will either guess or ignore them entirely.

‍

‍

How to communicate AI guidance without sounding like a killjoy.

‍

This is where tone really matters. If your AI guidance reads like it was written by someone who hates both technology and joy, adoption will happen anyway. It’ll just happen quietly and under the radar.

‍

Better approaches include:

• Plain English, not legalese
• Real examples instead of abstract rules
• “Here’s why” explanations, not just “don’t do this”
• FAQs based on real employee questions
• Clear signposting to where people can ask before using AI

‍

One simple rule of thumb that might be worth communicating: If you wouldn’t paste it into a public forum, don’t paste it into an AI tool.

‍

The big one: reassuring employees about job security.

‍

Let’s not dance around this. When organisations talk about AI, many employees start asking themselves one thing: “Is this going to replace me?” Ignoring that fear doesn’t make it go away. In fact, it usually makes it louder and fiercer. And considerably more widespread. Comms teams play a critical role in addressing this honestly and without making promises no one can keep.

‍

The most effective reassurance doesn’t sound like: “AI will never affect jobs.” No one believes that.

It sounds more like:

• AI will change how work is done but not eliminate the need for human judgement
• AI is intended to remove repetitive, low-value tasks, not people
• Human skills like creativity, empathy, problem-solving and context still matter
• The organisation is committed to upskilling, not replacing, its workforce
• People will be supported to learn how to work alongside AI

‍

Transparency builds far more trust than forced optimism.

‍

If there are roles likely to change significantly, it’s better to acknowledge that and talk about reskilling pathways than pretend nothing is ever going to change.

‍

‍

What good AI comms looks like in practice.

‍

In organisations doing this well, AI communication is:

‍

Ongoing, not a one-off announcement.
Employees get updates as tools evolve, policies change, and lessons are learned.

‍

Two-way, not broadcast-only.
People can ask questions, raise concerns, and suggest improvements.

‍

Role-aware.
What’s appropriate for marketing might be completely inappropriate for legal or finance.

‍

Consistent.
Leaders don’t say one thing while managers quietly discourage usage out of fear.

‍

Grounded in reality.
No hype. No doom. No paranoia. Just clear guidance and support.

‍

‍

A quick word on ethics, bias and trust.

‍

AI isn’t neutral. It reflects the data it’s trained on, which means bias is a real concern, particularly in areas like recruitment, performance assessment or decision support.

‍

Comms teams don’t need to become data scientists, but they do need to help reinforce a few principles:

• AI outputs should always be reviewed by humans
• Decisions that affect people should never be fully automated without oversight
• Transparency matters - employees should know when AI is being used
• Fairness and accountability don’t disappear just because a tool is clever

‍

If employees feel AI is being used on them rather than with them, trust will evaporate very quickly.

‍

‍

The internal comms opportunity.

‍

Yes, there is one! Handled well, AI is actually a gift to internal comms teams.

‍

It can help us:

• Draft faster and test ideas more easily
• Repurpose content across channels
• Personalise messaging at scale (carefully)
• Spend less time on admin and more time on strategy

‍

But our real value isn’t in using AI. It’s in helping everyone else use it wisely. We are the translator between technology, leadership, policy and people. That’s not a side role. That’s central.

‍

‍

The bottom line.

‍

AI in the workplace isn’t a future issue anymore. It’s a present one.

‍

Comms teams don’t need to have all the answers, but they do need to ask the right questions, set the right tone, and create space for honest conversation.

‍

Clear guidance beats silent fear.
Honest reassurance beats empty hype.
Human judgement still beats automation every time.

‍

If your AI comms helps people feel informed, supported and trusted rather than monitored or replaceable, you’re doing it right.

‍

And if you’d like help shaping that message without scaring the life out of everyone,  well… you know where to find us.

‍