Do You Ever Feel Like You Are the Only One Not Keeping Up With AI?

There is a feeling that a lot of knowledge workers are carrying right now and for a huge majority of them, they are not sharing it out loud. It sits somewhere between the stomach and the chest. It shows up when another large tech layoff is announced and you find yourself doing the quiet calculation: how far away is that from my organisation, my team, my role? It shows up in the half-second pause before you answer an AI question in a meeting or at a dinner, that internal flicker of: am I about to sound like the smartest person in the room, or the person with last week’s news? It shows up as a low hum of disappointment in yourself, you should know more. You should be reading more, learning more, tracking this more closely. Everyone else seems to have a view. Why don’t you have a clearer one?

I know that feeling because I had it. Not as an observer of other people’s careers, but in my own. I have spent nearly thirty years in digital transformation and organisational change, and for the first time in that career, I felt genuinely out of my depth with a technology I knew was going to affect everything I do and everything my clients face. So I did what I do when that feeling gets loud enough: I enrolled in a university AI Executive Programme. Not to become a technical expert, but to stop winging it on headlines. To actually know what I was talking about.

This piece is what I learned, not just from the programme, but from paying close attention to what is happening around all of us right now. And the first thing I want to tell you is this: the feeling you are carrying is not a personal failing. It is the entirely rational response to a situation that has been structurally designed to feel exactly this way.

The confusion is not accidental

Open LinkedIn on any given morning and you will find more AI content than you could read in a week. Some of it is grounded and genuinely useful. A significant proportion is speculative, commercially motivated, or generated by the very technology it claims to explain. And somewhere in all of that, you are trying to answer a question that actually matters to your life and your work: what is real, what is noise, and what do I actually need to know?

The difficulty is that you often cannot tell. This week I was called out on the internet by someone who assumed my writing was wholly AI-generated. I found that more clarifying than insulting – again, thank you Paul. It is a sign of how far the noise has travelled, and how hard it has become to trust what you are reading, including, apparently, writing that is demonstrably human.

The overwhelm you feel when you try to keep up is not a sign that you are not trying hard enough. It is the predictable consequence of an industry that has every incentive to keep the volume turned up, and very little incentive to turn it down.

Why the volume stays turned up: what is really driving it

In 2023, in an interview with The Atlantic, Sam Altman, the CEO of OpenAI, said:

“A lot of people working on AI pretend that it’s only going to be good; it’s only going to be a supplement; no one is ever going to be replaced. Jobs are definitely going to go away, full stop.”

Then at a Federal Reserve conference in 2025, he warned that entire classes of entry-level white-collar roles would vanish. Then in May 2026, at a conference in Sydney hosted by the Commonwealth Bank of Australia, he reversed. He was, he told the audience, “delighted to be wrong.”

In the same month, at an Anthropic financial services briefing in New York, alongside JPMorgan CEO Jamie Dimon, he reframed the argument entirely, suggesting jobs would transform rather than disappear.

"if you automate 90% of the job, then everyone does the 10% of the job, and the 10% kind of expands to be 100% of what people do and kind of 10x their productivity."

Two of the most influential voices in artificial intelligence walked back major claims in the same news cycle. Both companies are reportedly approaching IPOs.

Take a breath with that for a moment. Because this is not just a story about two men changing their minds.

The AI industry is competing simultaneously for the best engineering talent in the world, for research credibility that attracts further investment, for regulatory positioning that shapes the rules being written around them, and for the kind of sustained public attention that determines who gets to be central to this conversation for the next decade. But there is something else driving this that deserves to be called out. We have entered a new era of trillion-dollar valuations and an intensely competitive race to the top of the technology ladder. To name just a few: Altman, Bezos, Musk and their peers are not merely building companies. They are competing to be the defining figure of this century, to have their name attached to the technology that reshapes everything and the biggest dollar amount wins. The boldest claim gets the headline. The person who called it first gets the legacy. That dynamic does not encourage nuance or caution. It rewards whoever sounds most certain, most consequential, most indispensable to a future that only they have the courage to describe.

Bold, consequential claims are how you win on every single one of those fronts at once. If you are the person who understood what AI would do to society before anyone else did, you attract the researchers who want to work on something that matters, the investors who want to be part of something significant, the press coverage that keeps you in the conversation and wins the clickbait, and the policymakers who put you in the room when decisions are made.

Being first, and being dramatic about it, is a strategic asset. Nuance or indifference does not get you on the stage.

And then there is something Altman said that I find genuinely valuable. “I believe that so much of society here is going to be impacted by this, that we are all stakeholders, and it is better for us to be going in the direction of too much transparency and occasionally being wrong.”

Occasionally being wrong. It is a remarkable framing. It positions overclaiming as a feature of responsible leadership rather than a problem with it. Boldness in the service of preparedness. Transparency, even if imperfect, or grossly incorrect.

But here is what that framing does not account for. The everyday reader who absorbed the original fear and quietly restructured their career development around a warning that evaporated. The job seeker who feels like their future is doomed because the role they have trained for and spent their whole adult life doing is (apparently) completely redundant. The leader who made a workforce decision, because they thought they were behind, based on a signal that was then revised without a second thought. The professional who spent six months carrying the weight of a prediction that the person who made it is now "delighted" to have gotten wrong.

Ultimately, the fear reaches everyone, yet the correction reaches far fewer. And so far, there appear to be no consequences for the people making these predictions. The accountability lands on the people who believed them, not on the people who made them.

That is the system we are now navigating. Knowing that it exists does not make it less exhausting, but it does mean the exhaustion is no longer your fault.

What you can actually do: building your own reliable picture

The goal is not to keep up with everything. The goal is to build a small, reliable picture that you actually trust, and to update it as the evidence changes rather than as the noise fluctuates.

Here is what that looks like in practice.

Find two or three sources and return to them consistently. 

• MIT Technology Review (technologyreview.com) is one of the most dependable general sources available. It employs journalists rather than content producers, covers both the technical and human dimensions of AI, and does not treat every development as either catastrophe or miracle. A limited number of articles are free each month before you hit the subscription wall, and for most professionals that is enough to stay genuinely informed.

• The Rundown AI (therundown.ai) is one of the most widely read free AI newsletters available, with over two million subscribers and no paywall. It gives you headlines and practical how-tos in one daily email and is genuinely accessible for a non-technical professional.

• TLDR AI (tldr.tech/ai) is more technical in tone but free, published Monday to Friday, and covers research and developments that the more general sources often miss.

• The Verge’s AI coverage (theverge.com) is free to read, well-reported, and consistently frames AI through a human and societal lens rather than a purely technical one.

• The World Economic Forum’s AI tracker (weforum.org) gives a global view that most US-centric coverage misses.

• And when something significant is said at a major conference, go to the primary source transcript. Event hosts usually publish them, and they are almost always more nuanced than anything the secondary coverage produces.

What is not worth your time: newsletters that arrived without your active subscription, thought-leader content that makes sweeping predictions without citing methodology, and anything that leads with a jobs number without a methodology attached.

Run the incentive check before you react to a big claim. When something alarming lands, before you feel the fear or share the post, pause for a few minutes. Who said this? What do they gain from it being true? What do the people who study this for a living, rather than build it or sell it, say? That few-minute practice will save you more anxiety than almost anything else on this list.

Talk to the people in your organisation who are actually using AI tools. Not the people presenting the strategy. The people using the tools day to day. Ask them what is working, what is not, and what surprised them. That ground-level signal is more useful for your real decisions than almost anything coming from a conference stage.

Give yourself permission to remain curious without requiring a definitive view. You do not have to have a confident position on every development. Staying open and willing to update is not uncertainty as weakness. It is honesty as a professional discipline. The people doing this well are not the ones who sound the most certain. They are the ones who are most comfortable saying: I am watching this, I am forming a view, and I will tell you when I have one worth sharing.

Learn deliberately, not desperately

There is a version of AI learning that feels like panic reading: consuming more and more content in the hope that eventually you will feel caught up. It does not work. The content regenerates faster than you can absorb it, and the feeling of being behind only deepens.

The version that works is deliberate, bounded, and cumulative.

YouTube has excellent explainers that assume no technical background and are genuinely accessible for a professional with limited time. Google, Coursera, edX, and LinkedIn Learning have free and low-cost AI literacy courses that a working professional can complete in a few focused hours. Books are worth more than we tend to give them credit for in this era of constant short-form content. A well-argued book by someone who has spent years thinking carefully about these questions gives you something that no newsletter can: a framework for reading everything else.

If your organisation is running an AI pilot, a working group, or any kind of structured exploration, raise your hand to be part of it. You do not need to be the technical expert. You need to be present in the room where the real picture is forming, because that is where your most useful learning will happen.

And experiment personally. Use the tools. Notice where they help and where they do not, where they produce something genuinely useful and where they produce something that sounds fluent but is hollow. Your own firsthand experience is data that no amount of secondhand commentary can replace.

The programme I enrolled in did not make me an expert. It gave me a framework, a foundation, and the confidence to know the difference between what I understand and what I am still forming a view on. That turned out to be exactly what I needed.

What Altman admitted that almost nobody noticed

There is one more thing he said in Sydney, and I have not seen it widely picked up.

He mentioned that he had been using AI to manage his own emails and Slack messages. Then he stopped. He said he values the human interaction more.

The CEO of OpenAI personally decided that automating his own human connection was not worth it. He wanted to actually be in contact with actual people.

And it points toward something that matters more than any prediction about which jobs will or will not exist in five years. We are in a moment that is genuinely inviting us to ask what is irreducibly human. Not as a philosophical exercise but as a practical question about how we want to spend our time and attention.

If the machines take more of the routine and the repetitive, what do we do with what remains? I think the answer is to go further into what has always been irreplaceably human. Read the books that change how you think. Cook the food that requires your full attention. Make the art, listen to the music, move your body, tend the relationships that matter to you. This is not a retreat. It is a return.

A return to the parts of yourself that no algorithm can replicate: the creativity that surprises even you, the connection that reminds you why any of this matters. When you step away from the feed and into your own life, something tends to happen. You remember who you are beyond your job title and your LinkedIn profile. You discover, or rediscover, capacities you did not know were waiting. And those capacities, whatever form they take for you, have a strange habit of contributing to a different and richer outlook on everything else, including the future you are trying to navigate. You just never know what you might find there.

The feeling I want to leave you with

You are not supposed to know it all. The people making the biggest claims about where this is going do not know it all either. What I hope you take from this is not a reading list, though there is one above, and not a framework, though there is one of those too.

It is permission to stop measuring yourself against an impossible standard of total comprehension, and to start measuring yourself against a more honest one: am I investing enough attention, consistently enough, in the right places? Am I staying curious rather than defensive? Am I learning actively rather than anxiously?

If the answer to those questions is yes, even partially, even imperfectly, then you are doing exactly what this moment asks of you.

That is enough. It is genuinely enough.