Hi there!

Welcome to the 36th edition of Work in Beta.

In this edition, we take on the question of how to make your career safe from AI, and the answer is five things you can build inside the job you already have.

So, let's dive in!

At some point this year you've wondered whether AI will take your job, and what you should do about it. Most of the advice you'll find is one of two things. A list of jobs AI can't replace. Or a list of skills like adaptability and critical thinking, which sound nice and give you nothing to do on Monday. Neither helps, because there is no such thing as a safe job. AI doesn't take whole jobs. It takes parts of every job, including the ones on the safe list.

Here is what one person did about it instead. Kenya Solomon was a security guard in the US. When his company gave staff ChatGPT to try, he picked a problem nobody had asked him to solve. The fire-alarm panels in the buildings could only be read by a few experienced staff. He built a tool where a colleague photographs the alarm code and gets the steps to respond. The company created a technical project manager role for him as OpenAI tells it.

He didn't go looking for a safer job. He did two things instead. He picked a problem nobody had given him, and he made himself responsible for it. Those are two of the five things that make anyone hard to replace. The other three capabilities, you can build at the work you already do every week.

This is deciding what's worth solving before anyone asks you to. Solomon picked the alarm panels. Nobody assigned them to him.

AI does the work it's given and answers the questions you ask. It can't notice which problems are costing everyone time, and it can't tell you what to solve.

How to build this capability? Once a week, write down one thing in your team that keeps going wrong and that nobody owns:

After a month you'll have many items in this list, and one of them will be worth taking.

This is the history and the unwritten rules of your work: which vendor missed two deadlines last year, why the process has that extra step, why this client always wants the numbers a different way.

AI can't work any of this out on its own. It only knows what you give it, and most of us don't give it much. That's why the first draft AI gives you reads so generic.

How to build this capability? Every time you hand AI a task, add the context it doesn't have: the client's history, or the thing that went wrong last time. Then save that context in one file. After a few months, that file holds every exception, every unwritten rule and all the history you've picked up, and it becomes the one document in your company that nobody else could have written.

This is having your own view on a piece of work and being able to say why. It shows up in decisions: which vendor to go ahead with, whether to accept the client's discount request, whether to push the launch date. Someone with judgement has a view and can defend it. Someone without one forwards whatever came back.

AI makes this matter more, not less. Anyone can now get a sensible recommendation from AI in a minute, so a sensible recommendation on its own doesn't set you apart. What does is being the person who looked at it and said "no, and here's why", or "yes, and here's the risk". That's the person whose view gets asked for the next time.

How to build this capability? Say a client asks for a 15% discount to renew. Here is what you can do:

This is having your name attached to the outcome. If it goes wrong, people come to you. Take the client renewal again. Whatever discount you recommended, if the client leaves, they would ask you. Nobody would blame AI and nobody fires AI. A company still needs a person who is responsible for the result, and that's the person it keeps.

How to build this capability? Go back to your list of problems nobody owns and pick the one that matters most. In your next team meeting, tell your manager you'll own it, and then see it through until it's done. This is the move that changes how people see you, and it needs no promotion to start. Most people don't do this because it's a risk. If you take on something nobody asked you to and it doesn't work out, everyone will know.

This is being comfortable enough with AI to use it on your own work every day, and knowing what it does well and where it lets you down. You don't need to code, and you don't need prompt tricks. You need to have used it on enough of your own work to know what to expect from it.

How to build this capability? Pick one task you do every week and do it with AI every single time, even in the weeks when doing it yourself would be quicker. Each time, notice where it let you down, and give it what it was missing the next week. After a couple of months you'll know exactly what to expect from it on that task, and that knowledge carries over to the next one.

Here is a simple way to see where you are on the five. For each row, note whether you did it in the last month, and if you did, write down one example. It takes five minutes, and the point is to be honest with yourself rather than to score well.

Fill it in this month and keep it. Then fill it in again next month, and the month after. The first time, the empty rows will show you which of the five you've never worked on. By the third month, you will notice the difference. Keep going after that too. AI keeps getting better, so this isn't a one-time exercise.

There is no job that AI can't touch, so stop looking for one. What you can do is build the five things that make a person hard to replace, and you can build all of them inside the job you already have, using the work you already do every week. Solomon did it as a security guard. None of the five depends on your title.

Most professionals know AI can do more for them. The gap isn’t awareness - it’s knowing where to start, what to change, and how to make it stick.

That’s what we work on through Work in Beta.

For individuals, we run working sessions, not teaching sessions. You bring a real problem from your actual work; we build the solution with you, live. You’re borrowing our learning curve instead of grinding through your own. You walk out with something that works and the muscle to keep going. When you get stuck later, we’re a message away.

You probably have a version of at least one of these:

Bring that. We’ll build it with you.

For organizations, AI adoption is a people problem, not a technology problem. Your teams have the tools, what’s missing is the translation layer between AI capability and daily work. Which processes to redesign, which habits to break, how to build genuine fluency, not just awareness. We help close that gap through hands-on training, process redesign, and deep adoption engagements. Not advisory, forward-deployed.

If any of this resonates, email us at [email protected] / [email protected] and we will figure out how to work together.

You can reply to this newsletter email if you want to share something specific with us. Have a great week ahead and see you again next week!

-Sonali & PD