Getting Paid for AI Skills: What to Do When AI Work Lands in Your Job but Not in Your Title or Pay

Nov 3

11 min read

Updated: 4 days ago

Woman with curly hair reviewing a highlighted job description at her desk, planning for getting paid for AI skills

Erica Anderson Rooney is an ex-CPO turned executive coach and AI educator on a mission to get more women into positions of power and keep them there. She is the author of The AI Gap (2026), host of the Glass Ceilings & Sticky Floors and AI Voice or Victim podcasts, and NC Chapter Lead for Women Applying AI. Her SNAP and AI Continuum frameworks are taught to thousands of women through CHIEF, WICT, and Fortune 500 leadership programs. She is the sought-after speaker at the intersection of women, AI, and leadership.

This is how it always starts… with one small favor. Your manager pings you on a Tuesday: "Can you show the team how you built that client summary with AI?" And the next thing you know, by Friday, you are fixing everyone's prompts, building the template the whole department now uses, and answering questions from people two levels above you. Congrats girl, you are officially "the AI person on the team."

But here is the deal…nobody changed your title. Nobody changed your pay. Nobody even made an official note of it. And that is exactly why I want to talk about getting paid for AI skills today, because a new 2026 report just put data behind something I have been watching happen to women all year (if not my whole career, just copy/paste different skill.)

And what high achiever do you know that hasn’t had this experience? Where do you think this running joke came from: “Do good work and you’ll be rewarded with MORE good work?” Women know this all too well.

This is the pay gap being built in real time, y'all. And I do not want you finding out at review season that your AI work never made it onto paper. I do not want you to, yet again, miss out on the real pay raise you deserve.

The Newest Invisible Job at Work

We all know the old version of this. Somebody takes the notes. Somebody plans the team lunch, onboards the new hire, remembers the birthday card and quietly mentors the junior people. That somebody is usually a woman, it may have been you, and that work almost never shows up in your paycheck.

Now there is a new version that involves AI. She trained herself on AI. Paying for the course out of pocket, and staying up late to learn the tool. Then, at work, she tests the new tech before anyone else. She writes the prompt library. She trains the team, fixes it when it breaks, and translates what the AI did for the executives who will never open it themselves. It is skilled work. It saves the company real hours and real money. And it lives nowhere on record. She does it because she works hard, wants to be on top of her game, and yet… it doesn’t translate to pay.

Sound familiar?

Women end up here for good reasons. We see the gap and we fill it. We like being useful. We figure the work will speak for itself. (Girl, if only.)

The trouble is that "the AI person on the team" is shaping up to be the newest kind of unpaid work, and it is landing at the exact moment AI skills are becoming some of the most valuable skills in the job market.

What the Data Says About Getting Paid for AI Skills

On October 14, 2026, Oyster, a global employment platform, released its 2026 Global AI Talent Report. Oyster surveyed more than 1,000 HR and people leaders in the United States, the United Kingdom, Canada and Europe in August 2026, and analyzed more than 15,000 job descriptions posted through its platform since 2022.

The headline number: 95% of HR professionals say employees at their company use AI tools every day in roles where AI is not part of the official job description.

Ninety-five percent is a big number, y’all. So, here we have almost everyone doing the work. And I wanted to share that stat, because this is where it gets personal. Among companies that added AI requirements to existing roles, 62% changed neither the title nor the pay. And 46% never formally told the people doing those jobs. (Sneaky, right?)

You mad yet, girl? Your job changed, your company knows it changed, and nearly half the time nobody acknowledged it. They didn’t want to give you an opportunity to ask for that raise that you so rightfully deserve.

And it keeps going. When an employee automates part of her own job, 38% of HR professionals say she would be rewarded. But 24% say she would simply be given more work. (Well, here is that joke, good work gets rewarded with MORE good work.) So roughly one in four HR professionals expect your new AI skill to earn you a heavier plate. Love that for us.

And for job seekers, 68% of companies now ask candidates what they have personally built with AI. Of those companies, 53% say the answer does not affect hiring or pay decisions. They want to hear the story, and then the offer stays the same.

The demand is real, too. Oyster found that AI skill requirements in the roles it hired for tripled from 5% in 2022 to 15% in 2026, and nine in ten of those roles do not name AI in the job title. The skill is being asked for, and the title is lagging behind.

One honest note before we go further. This is a company report, released as a paid press release, and Oyster sells hiring and payroll services. Its survey was run through the research platform Prolific, and Oyster itself says its platform data reflects its own globally minded customers, not the whole market. So I treat it as a strong signal and read it alongside other research.

The thing is, it lines up with what other research keeps finding. PwC's 2026 Global AI Jobs Barometer, released in June 2026 after analyzing more than one billion job ads, found that workers with AI skills command an average wage premium of 62%. (I dug into that number and what it means for women in The AI Skills Wage Premium Is 62%.) So the market is paying for AI skills. Your company just might not be paying you for yours.

And women are starting from behind. In McKinsey and LeanIn.Org's Women in the Workplace 2025 report, only 21% of entry-level women said their manager encouraged them to use AI, compared with 33% of men at the same level. A Lean In survey reported by Axios in April 2026 found that among people who use AI, 18% of women had been praised for it, compared with 27% of men.

Less encouragement. Less praise. And now, AI work that never gets written down. (That last piece did not surprise me, but I live in this work.)

Why Unwritten Work Gets Skipped at Raise Time

When I was a Chief People Officer, pay conversations ran on paper. The job description. The leveling guide. The performance review. The manager's notes. (If the work was not in one of those places, it was very hard for anyone to defend paying for it, no matter how much everyone loved you.)

That is what makes the Oyster numbers so dangerous. Unwritten work is the easiest work to overlook when raises get decided, and AI work is going unwritten at a massive scale.

I once saw a post in a working moms group from a woman who had spent six months doing extra work, hoping it would lead to a promotion or a raise. Six months! So I asked her one question. Had she ever talked with leadership about whether that direction was something they even wanted?

She had not. And I said something on my podcast, Glass Ceilings & Sticky Floors, in my episode with Dr. Cynthia Bentzen-Mercer, that I still believe with my whole heart: you can never do extra work thinking it's going to lead to a promotion. The work has to be seen, named and agreed on, or it stays a favor.

That is a sticky floor. Keeping your head down and hoping someone notices feels safe, and it keeps you exactly where you are. AI work makes that floor stickier, because so much of it happens quietly in a chat window where nobody else can see it.

How Do I Get Paid for the AI Work I Already Do?

Woman with silver-blonde hair and glasses showing a colleague a workflow on her laptop in a meeting room

Get your AI work written into your job description and your review goals, keep a record of what you built and what it saved, and bring that record into one conversation about title and pay. Companies pay for work that is written down and measured. Your job is to get your AI work on paper before review season.

Here are the four moves I would make this month.

Move 1: Ask for Your Job Description to Catch Up With Your Job

Pull up your official job description. Most of us have not looked at ours since the day we were hired.

Now highlight everything you do today that is not on it. Then write three to five new lines yourself, in plain business language. For example: "Builds and maintains AI workflows for the team's monthly client reporting." "Trains team members on approved AI tools and reviews AI-generated work for accuracy." "Owns the department's prompt library."

Bring those lines to your manager and say something like this: "My role has changed a lot this year, and I want my job description to reflect what I am doing now. I drafted a few lines. Can we update it together, and can we talk about what this means for my level?"

Writing the draft yourself matters. The person who writes the first draft shapes the final version. And if your manager says HR owns job descriptions, ask who in HR you should send it to. (When I was in HR, a clean draft from an employee made my job so much easier.)

Move 2: Put Your AI Work in Your Goals and Your Review

Even if the job description takes a while to change, your goals can change this quarter.

Ask to add one goal tied to your AI work. Something you can measure, like "Cut the monthly reporting cycle from five days to two using AI" or "Train eight team members on our approved AI tools by March."

Then, in your self review, give your AI work its own section. Name what you built, who uses it and what changed because of it. And ask your manager one question in your next one-on-one: "How will this work be weighed in my review?" If the answer is fuzzy, that tells you exactly what you need to fix before calibration.

Move 3: Keep a Build Log, Because Receipts Matter

This is the record that turns into your negotiation. Start a simple running note today. Every time you build something with AI or change how work gets done, add one line with four things: what you built, who uses it, the time it saves, and the problem it solved.

It can look like this. "Built an AI first draft process for client proposals. Used by six people. Saves about three hours per proposal. Proposals now go out two days sooner." That one line is worth more in a pay conversation than "I've been really busy with AI stuff."

Count the hours too. I wrote about how to track and claim the time AI saves you in AI Time Savings at Work, and the build log is the other half of that same story. The hours show what AI saved. The log shows what you created. Together, that is creating the paper trail.

Keep the log somewhere you own, like a personal note on your phone, so it goes with you if you change jobs. Remember, 68% of companies now ask candidates what they have built with AI. You want that answer ready, with numbers.

Move 4: Bring the Title and the Pay Into the Same Conversation

Woman in a cream blazer discussing her title and pay with her manager across a desk in a glass office

Once your work is written down and your log has a few months in it, ask for the conversation. Do not split it into two. Title and pay belong together, because a new title with no raise is just more responsibility with a nicer label.

Here is a script you can make your own: "Over the last six months, my role has grown to include building and running our team's AI workflows. Here is what I built, who uses it and what it has saved. AI skills are commanding a real premium in the market right now, and I would like to talk about a title and compensation that match the job I am doing today."

If the answer is "not right now," ask three follow up questions. What would need to be true for this to happen? When can we revisit it? Can we put that in writing? Then send a short email recapping the conversation the same day.

If your hands get sweaty just reading that script, you are in good company. I walk through how I handle the fear that shows up in pay conversations in my podcast episode How I Negotiated a $100K Raise With the SNAP Method. The fear is normal. Let it ride along, and ask anyway.

If You Lead a Team, Write It Down for Them

Now, managers and people leaders, this part is for you. You can fix this from your side of the desk too.

Look at your team this week. Who has become the AI person? Who is quietly training everyone else? Then look at their job description. If the two do not match, you have found your first project.

Update the role before review season. Add an AI goal to the people doing the work. Name their contributions out loud in team meetings and in writing to your boss. And remember that Oyster found 46% of companies never told employees their jobs had changed. Be the leader who tells them, and then pays them.

Watch the pattern, too. If the AI work keeps landing on the same few women, and the credit and the stretch assignments keep landing somewhere else, you have a pay equity problem forming right in front of you.

Whoever Writes the Job Description Shapes the Pay

Here is the bigger picture. Pay gaps get built from thousands of small decisions. Whose work got written down. Whose work got named in the meeting. Whose new skills got a title, and whose got a thank you.

AI is a power shift, and right now job descriptions are being rewritten quietly across almost every company. If women do not make sure our AI work lands on paper, we will look up in three years and find the same old gap with a brand new cause. (I broke down where the pay gap stands in 2026 in Is the Gender Pay Gap Getting Worse in 2026? over on my other site.)

This is part of the women and AI gap we are working to close, and it is one of the most fixable parts. The fix starts with a document.

What I Want You to Take From This

If AI is in your job, it belongs in your job description, your goals, your review and your pay. Start with the document. Keep the log. Ask for the title and the money in the same breath.

And if you lead people, give your team the same protection you want for yourself.

If you want a room of women working through exactly these conversations, come join us in HER Collective. Every week we meet live for group coaching, we bring in experts for masterclasses on topics like negotiation and pay, and we do hands-on AI training for women who do not work in tech. Bring your build log. We will help you turn it into your ask, and we will practice it with you before you walk into that one-on-one. We saved you a seat.

I am cheering you on, and I want your next title to match the work you are already doing.

Erica

Frequently Asked Questions

Should I get a raise for using AI at work?

If AI has changed what your job requires, it is fair to ask. Oyster's 2026 Global AI Talent Report found 62% of companies that added AI requirements to existing roles changed neither the title nor the pay. Get the work written into your job description first, keep a record of what you built, then ask for title and pay together.

How do I add AI skills to my job description?

Pull up your current job description and mark everything you do now that is missing from it. Write three to five new lines yourself in plain business language, such as “builds and maintains AI workflows for team reporting.” Bring the draft to your manager, ask to update it together, and ask what the change means for your level.

What is a sticky floor, and how does AI work make it worse?

A sticky floor is Erica Anderson Rooney's term for the habits and beliefs that hold women in place, like keeping your head down and hoping good work gets noticed. AI work makes it stickier because much of it happens quietly where nobody sees it. Writing your AI work down and asking about it gets you unstuck.