AI: The Reputation Signal · Issue 7
The ladder didn't get automated. It got taken down
Elif Güvençer ·
"This is the way we do things in this part of the world."
I heard that sentence more than once when I first moved from Istanbul to London years ago and I still remember how it made me feel.
It stayed with me because of what sits underneath it. If you're a minority, or you're not from here, or you don't fit the shape of the room, it's hard. In communications it's harder. You can have the expertise and still struggle to make it count, because first you have to get past everything that says you don't belong.
I dropped the special characters from my surname so I could blend in. I wanted equal footing before my name gave me away. But when the spelling didn't give me away, my accent did. My expertise had to climb over any and all preconception every time. Less room for mistake. Ignore the comments, push through. A shout out to Rebecca, VP of HR at RepTrak. The special characters don't hold much power over me anymore, but she's the kind of inclusive HR leader who noticed they weren't showing up in the system and got it updated.
(Koray Camgoz has a similar story with the special characters in his last name).
So lately I've been thinking about the graduates of today and what AI is doing to the bottom rung they're trying to reach. I recently ran a masterclass for the Taylor Bennett Foundation PR graduates. We talked about how to build a career AI can't automate.
It made me look hard at the evidence.
First question: is it actually AI?
There's a lot of fear in the air. AI is coming for the jobs, the entry-level ones first. Some of that is true. Some of it is AI taking the blame for cuts that were coming anyway.
The most reassuring study I've read is the broken ladder research, by Peter Lambert and Yannick Schindler. The research is built on 243 million hires and 407 million online job postings, collected across the US, UK, Canada, and Australia during 2017-2025. It found that when you measure GenAI exposure and remote work separately, each looks like it's hollowing out junior hiring. Estimated jointly, remote work effect remains, while GenAI coefficient attenuates sharply and is often statistically indistinguishable from zero.
According to the Federal Reserve Bank of New York, remote work could account for as much as 64% of the overall rise in youth unemployment since the pandemic.
Juniors learn by being near people, especially in Communications. The overheard call. The edit you watch happen over someone's shoulder. Remote took that away. So in a way, the ladder wasn't automated. It was taken down.
(Ben Verinder has an interesting view about PR roles worth reading here).
When entry level jobs decline, who is impacted the most?
New YouGov research for the Taylor Bennett Foundation found two-thirds of UK adults expect AI to cut entry-level jobs. But ask who gets hit hardest and only 6% name young people from ethnic-minority backgrounds, the lowest of any group on the list that includes disabled young people, neurodivergent young people, young people from low-income households....The public can see the door closing. They can't see who gets pushed further to the end of the line.
There is also the uncomfortable reality of the door now being guarded by technology. In the Mobley v. Workday case, an applicant submitted a job application at 12:55am and was rejected by 1:50am. No human was awake to read it. That was a machine veto. I’ll park this topic for another post.
The majority of readers here aren't graduates but many of you are leaders and hiring managers on the other side of the table. It's worth thinking about what that means for entry-level PR roles.
I wrote The Two Clocks Framework™ for top of the funnel; for Communications leaders to think about what do they need to do about AI but it works as well at the start of a career as well. Which is what I centred the masterclass around.
The immediate clock: be useful today
This is reputation management in an AI-mediated world. Monitoring, content, the day to day operations.
We ran an exercise and built the files needed to personalize your AI so it sounds like you. Do that with the corporate brand voice and you're useful on content creation from week one on the job.
Then find the pain point. What does your manager complain about most? What's the repetitive task eating their week? Attempt to build the thing that fixes it. You don't need to know how to build. You need to know how to ask AI to teach you.
Make it concrete. A breaking topic lands and the monitoring agency hasn't touched it yet, so you build the ad hoc briefing before anyone asks. Reddit keeps showing up in AI answers, so you build an agent that watches the threads, then walk into your manager's office with: here's what people are saying, do we have anything that answers it, should we be in that subreddit. The tasks AI can take off your plate, you take, and you do them smart enough to be the junior people remember.
The structural clock: become irreplaceable tomorrow
This is what Communications needs to become to assume the role it needs to play in the AI era.
Ring-fence 10% of your time. Not to learn another tool, but to become the person AI can't replace.
One way to build the judgement for all of it: use what you already know. We all have a subject we're deep in. Football, gaming, whatever it is. Ask AI to do something in that area and watch where it gets things wrong. That's the first step in training your eye to spot where it's wrong everywhere else.
We did an exercise on judgement in the AI era. Imagine Marketing wants to launch an AI agent that speaks for the company, and it lands on your comms manager's desk for a view on the risk. Name three ways it could blow up the company's reputation. Name where the agent must never go. Then find the one question legal and IT would never think to ask.
The graduates I taught at Taylor Bennett shouldn't have to work extra hard and one day I hope the door is simply held open for them and that becomes “how we do things everywhere in the world.”
Until then, the advice I would follow if I were a graduate is this:
- Clock starts to run before the interview. Customize your CV, map the hiring manager/role, simulate the interview with AI.
- Walk in the interview with two or three real examples of how you can use AI in the role.
- When you land the role, be useful fast. Find a pain point, a manual task and automate it.
- You don’t have to know how to build things, you need to learn how to use AI to help you build things.
- Protect your own reputation: No shadow AI and check your facts.
- Learn what good AI looks like. Use a hobby to start.
- Ring-fence the 10% for you. Spend it on who you're becoming, not the next tool.
- Watch out for cognitive offloading. Never surrender the judgement call.
- Be demanding. Shadow your manager, ask to learn the craft AI can’t replicate.
- Use reverse mentoring to your advantage.
Graduate roles aren't a cost to trim. Holding the front door open is not only the right thing but also serves the function’s self-interest. Communications industry needs to re-think what an entry level PR role looks like and how to teach the craft in the age of AI. Otherwise it will see itself automated out of existence.