How to Show Your Employer You Are Ahead

The person nobody listens to

Most workplaces now have one. The colleague who mentions AI in every meeting, forwards articles nobody opens, describes tools rather than results, and is faintly evangelical about all of it.

They are usually enthusiastic and often right about the technology. They also have almost no influence, because they have made AI their identity rather than their method. Once colleagues file you under enthusiast, everything you say gets discounted before it is heard, including the parts that are correct and useful.

This matters, because you do want your employer to know you are capable here. The question is how to make that visible without buying the whole costume.

The answer is unglamorous. Talk about outcomes and risks, not tools. Do it rarely and specifically. Let other people be the ones who say the word.

Show a before and after, not a demo

A demo shows what a tool can do. Nobody in your organisation is short of demos, and they all look impressive in a way that does not survive contact with real work.

A before and after shows what happened to a real task in your real job. It is much less exciting and vastly more persuasive, because it is checkable.

Pick one recurring task you genuinely improved. Write down how long it used to take, how long it takes now, what the output looks like in both cases, and what you still do by hand. Include the last part honestly. Saying that you still verify the figures yourself makes the whole account credible in a way that a claim of pure speed never does.

Help me write a plain, unexcited before and after for one task at work.
Structure: what the task is, how it used to be done and roughly how long it took, how it is done now, what I still do manually, and one honest limitation.
Rules: no adjectives like transformative or game changing. No recommendation that anyone else adopt it. Under 150 words. It should read like a status note, not a pitch.
MY DETAILS:
[describe the task, old process, new process, what you still check, what does not work]

Offer to write the team's approach

Here is the most underrated move available to you.

In most teams, people are already using these tools quietly, and nobody has said what is allowed. Nobody wants to ask, because asking might produce a no. So shadow use grows, nothing is written down, and the first time it becomes a topic is when something goes wrong.

Offering to draft a short, practical team approach solves a real problem for your manager and positions you as the sensible person on the subject. Not the expert, the sensible one. That is a better position.

Keep it short and boring: what we use these tools for, what we do not put into them, what always gets checked by a person, and who to ask when unsure. One page.

Help me draft a one-page practical guide for how my team uses AI tools. Keep it short, boring and usable, not a policy document.
Cover: what we find them useful for, what should never be pasted into them, what always needs a human check before it goes out, how we note when AI was involved in a piece of work, and who to ask when someone is unsure.
Write it in neutral team language, not as my personal opinion. Flag anywhere I would need to check my organisation's own rules first.
MY TEAM CONTEXT:
[describe the team, the work, and any rules you already know about]

Be the one who flags the risk early

The other half of visible competence is being the person who spots a problem before it lands. Not the person who blocks things, and not the person who worries aloud in meetings. The person who says the specific thing at the point when it can still be acted on.

That means being concrete. Not that AI output can be unreliable, which everyone knows and nobody can act on. Instead: this particular section rests on a figure we have not verified, and if it is wrong the client will find it in the first review.

Raise it early, privately where possible, and bring the fix with you.

I want to raise a risk about how AI is being used on a piece of work, without sounding like I am blocking it or lecturing anyone.
Help me phrase it so it is specific about what could go wrong, includes a practical suggestion, and is short enough to say in a meeting or send in one message.
Give me two versions: one for raising it before the work goes out, and one for raising it when the work has already gone out.
THE SITUATION:
[describe it]

Do those three things over a few months and something quietly useful happens. You become the person your manager checks with on this subject, without ever having claimed to be an expert on it. That is a considerably stronger position than the one the evangelist holds.