AI Didn't Come for Your Job. It Came for Your Excuses.
A few weeks ago, I closed the AMEC Global Summit in Dublin. AMEC is most commonly known as the International Association for the Measurement and Evaluation of Communication, "a leading global professional body for PR and communications research."
The sexy and not so sexy part of comms, I know.
I was standing in front of a room full of communicators, researchers, measurement professionals, agency leaders, and SaaS platform builders -- the people who have spent careers being told by some that data and analytics are a "nice to have" or squishy.
These are the people who have been gathering evidence, tracking effectiveness, and making the case for comms' strategic value for decades, often without the budget, the right tools, or frankly way that commands the respect it deserves.
I told the room that this is a Revenge of the Nerds moment for us all, and it's time to claim it.
The timing was kind of precious. Just days before I took the stage, Open AI's Sam Altman told an audience in Sydney that he was "pretty wrong" about AI's impact on white-collar employment. "I thought there would have been more impact on entry-level white-collar jobs being eliminated by now than has actually happened," he said. "I'm delighted to be wrong about this."
Anthropic's Dario Amodei, who once warned that up to half of all entry-level white-collar jobs could dissolve within five years and unemployment could hit 10–20%, has since reframed automation as a multiplier of output rather than a destroyer of jobs. Both are preparing blockbuster IPOs. I'll leave the motive question to others. But the substance of the walk-back is right.
My keynote was called: AI didn't come for your job. It came for your excuses. I opened with a question: what have we been doing with our time that we couldn't defend even before AI arrived?
Our industry has hidden behind the same excuses for years. We don't know if a journalist will actually be interested in a story. We can't prove that our announcement actually increased shareholder value. We don't know how a crisis statement we issue will mitigate or accelerate media coverage or social chatter. We could not predict the next crisis or trend. We have long been getting paid for outputs and activities, not business outcomes.
That last one is the most honest thing the industry has ever said about itself, and it is the excuse AI will make completely indefensible. Now that AI produces most of the outputs and the activities, what's our intrinsic value? The outcome! Which means the industry needs to get comfortable paying for it, pricing for it, measuring it, and staking its reputation on it. That's not a threat. That's the job we always should have been doing.
AI did not arrive from nowhere. It is not a rupture or revolution. It is the latest chapter in a story that began fifty years ago, and understanding that changes everything about how we should feel about it.
1975: The PC democratized computing. For the first time, individuals had direct access to processing power. Everyone said it would eliminate jobs. Instead, it created entire new categories of work that hadn't existed before -- software developers, systems administrators, desktop publishers, IT departments.
1989: The web connected everything. Information became a network rather than a library. Again, the obituaries were written. For newspapers, for retailers, for travel agents. Some didn't survive. But the overall volume of human economic activity didn't shrink, it expanded, dramatically, into categories nobody had anticipated.
2006: Cloud computing scaled access. You no longer needed to own infrastructure to use it. The concern was that enterprise IT would collapse. What actually happened was that startups could now build at a scale previously reserved for large corporations, and an entirely new layer of the digital economy emerged.
Broadband and mobile apps then supercharged it all.
2022: AI made software intelligent. It stopped being a tool you operate and started being a delivery system that learns, adapts, generates and re-generates.
Each of these waves felt at the time like a civilizational disruption. Each one eliminated some roles, restructured others, and expanded the overall scale of human activity in ways that the pessimists couldn't admit or see coming. The pattern is not ambiguous. It has repeated, with remarkable consistency, across half a century.
What's different this time is not the direction of travel. It's the speed. AI is compressing into years what the PC, web, and cloud each took decades to accomplish. The pace is genuinely unprecedented. But the pattern -- new infrastructure creates new demand, new demand creates new roles, early movers capture disproportionate advantage -- that part is not new at all.
Without the first three waves, the fourth is impossible. AI is not a standalone breakthrough. It is the product of fifty years of accumulated infrastructure investment. Every GPU running a language model today exists because of the PC revolution. Every AI system trained on internet data exists because of the web. Every model accessible via API exists because of the cloud. We are not at the beginning of a technology cycle.
We are at the inflection point of one that has been building since Gerald Ford was in the White House.
That context should make you clear-eyed. The organizations that moved early in each of the previous waves didn't do it because they had certainty. They did it because they understood that the cost of waiting was higher than the cost of being wrong.
65% of Google searches now end without a click.
44% of users rely on AI-generated summaries as their primary source of information.
Generative AI reclaims 26–36% of communications time. Agentic AI? Up to 47%.
This is the current state of the information environment your communications operates in. The delivery system has changed. If your strategy was built for the old one, it needs to be rebuilt, not tweaked.
One more data point worth sitting with: Ipsos surveyed 21,000 people across 21 countries and found that most of the world registers AI as opportunity. Western markets register it as threat. That asymmetry is not accidental. Markets with less institutional legacy and fewer entrenched business models to protect tend to see new technology as additive. The communicators most excited about the AI-era playbook are not primarily in London, New York, or LA. They are in markets that cannot afford to wait.
I've written before about the JEEPS framework: Judgement, Experience, Expertise, People, Strategy -- as the answer to "what do I still have that AI doesn't?" I won't relitigate it here. The short version: the calls no algorithm can make are yours. Heuristics! Protect them.
The risk isn't replacement. It's "work slop" -- volume over quality, at the cost of the profession's credibility. Don't let AI make you a content factory. Let it make you a better strategist.
Junior roles are being cut back. That is one of the most strategically short-sighted decisions this industry is making right now. Junior communicators are not just output. They are the training ground for the senior judgement this industry depends on. You cannot skip the apprenticeship and expect experienced leaders in five to ten years.
The business model conversation can no longer be deferred either. Brands still buy hours. Agencies still sell them. That transaction will not survive the next two years in its current form. The firms that reprice around outcomes will thrive-- decisions made better, crises avoided, reputations protected. The ones that don't will be disrupted by the ones that do.
Most of the industry is still stuck between curiosity and experimentation. Curiosity is not a strategy. A Microsoft co-pilot wrapper on an existing workflow is not transformation. The window to actually move is still open. But it is not open indefinitely.
If the prevailing narrative around AI has left you anxious, I'd offer this reframe: you are not watching a disruption unfold. You are living through an infrastructure upgrade, one that follows a pattern we have navigated before, and one that has always, eventually, expanded opportunity rather than contracted it.
Being an AI optimist is not a posture. It is a conclusion you reach when you actually look at the history. Every wave of computing infrastructure — the PC, the web, mobile, the cloud — produced more work, more demand, and more value than the one before it.
The professionals who grasped that early didn't just survive each transition. They defined what came next. That is the position available to communicators right now. Not survival.
What is your real value proposition in an AI era? Not the pitch version. The honest one.
What is your moat and are you leaning into it, or just talking about it?
What does your organization look like in 24 months? Not aspirationally. Structurally.
The future of this industry is different. Not dead. Optimism is a choice. Unlocking it requires reflection, ingenuity, and action.
