Trent Gillespie
June 04, 2026
More than three years into putting AI inside real companies, we have not had a single client lay off a single person because of it. Not one.
I lead with that because the people who spent three years predicting the opposite just started walking it back.
Last week Sam Altman, who warned in 2025 that AI would gut entry-level white-collar jobs, said he was "pretty wrong." Dario Amodei, who once put the number at half of all white-collar jobs, now says AI is more likely to expand the work people do than erase it.
I have said the same thing from stages for years, so I will say it plainly here. The jobs apocalypse was never coming. And the layoff headlines you have been reading do not show what you have been told they show.
So far, companies have announced close to 50,000 job cuts this year that they pinned on AI, about 17% of all layoffs in 2026.
But that doesn’t mean it is true. Deutsche Bank has a name for the game: "AI redundancy washing." A layoff driven by over-hiring, higher interest rates, economic uncertainty, or pressure to look lean for investors sounds a lot better to Wall Street when you blame the robot. Most of those cuts were coming anyway. AI just made the better headline.
Strip the washing out and the ground looks a lot more steady. Yale's Budget Lab studied the labor market from 2022 through 2025 and found the mix of jobs people actually hold has barely moved since ChatGPT arrived.
The pain that is real is concentrated, however. Workers aged 22 to 25 in the most AI-exposed fields like software development are getting squeezed, and we should take that seriously. But a narrow slice of the workforce is being sold to you as the whole future of work.
Here is what I see in the rooms I am actually in.
Every leader I work with wants the same handful of things. Do the work better. Serve customers better. Grow. Get ahead of where their industry is going.
None of that shrinks a company. All of it needs people.
Let me put a number on it.
I mapped a mid-sized company recently. I looked at everything they should be doing to perform at their best with the tools sitting on their desks today. The number came to 374 processes.
They run about 75 of them right now. By hand. That leaves 299 they have not built yet.
Sit with that. The 299 do not exist. The 75 that do are not even automated. Someone has to build all of it. Someone has to design the workflows, apply the technology, check the output, and make it better over time. That is years of real work. It makes the company more capable, lifts the quality of what everyone produces, and leaves the business far more solid than it is today.
Will there be job losses along the way? Yes. In some places, of course.
But will there still be far more work than people to do it, for a long time? Absolutely.
The thing to watch is pace. Your industry is adopting AI right now, with or without you. The companies that move fast enough to stay relevant will have all the work they can handle.
One important note, however: if you are a business of one or a few, the risk is higher and I don’t want to understate that. A solo REALTOR, a single-shingle wealth advisor, or a small marketing agency often has no cushion if someone faster shows up.
Some of these operators will lose out, and not to AI itself. They will lose to the competitor across town who uses it to serve twice the clients, and to the tech platforms racing to automate the service outright.
The work does not disappear. It moves to whoever serves the customer best, and the tools to get there are finally within reach of a one-person shop. Be that person.
One of the best ways for many businesses to start is called ‘Speed-to-Lead’: how fast you reply to new leads you receive. The data on speed is brutal. A widely cited Harvard Business Review analysis of millions of leads found you are roughly 7 times more likely to qualify a lead if you reach them within the first hour than if you wait longer. That window is impossible to hit by hand on every lead, around the clock. It is exactly the kind of work to hand to software.
Here's what an AI agent handling your leads does:
Responds to every new inquiry within minutes, day or night, with a real answer instead of a generic auto-reply
Pulls the lead's source, history, and what they asked about, then tailors the message to it
Books the meeting or hands a hot lead to the right person while the interest is still warm
Follows up on its own across email and text on a set cadence until the lead replies or opts out
Logs every touch, flags the ones worth a human call, and reports what is actually converting
You can build much of this yourself with a variety of off-the-shelf tools like Zapier, Lindy, or n8n. Platforms like GoHighLevel, which a lot of agencies and local service businesses already run on, can text a new lead back the second it comes in, answer questions, book the appointment, and keep following up across text and email. There are hundreds more startups in the space. Of course, at Stellis AI, we can build it for you too.
Most businesses let good leads go cold. Not because they don't care. Because they never had the capacity to work every one.
Now they do.
The doom is softening for a reason. Here is how I would act on it this week.
1. Read layoff headlines as press releases, not forecasts
Most of the "AI layoffs" in your feed are cost decisions wearing an AI costume. Do not let someone else's redundancy washing set your strategy or rattle your team.
2. Inventory the work you are not doing
Your growth is hiding in the equivalent of those 299 processes, not the 75 you already run. List the customers you cannot serve, the offers you cannot get to, the follow-up that never happens. That list is your real AI roadmap, and not one item on it is a job to cut.
3. Build the bench to run it
The constraint is not the model. It is people who can put it to work. The companies pulling ahead are training their teams to design and run these processes right now. The most accessible AI has ever been still does not run itself.
The opportunity is already inside your building. It is everything you keep meaning to get to and never have the capacity for. That is where your attention belongs this week.
Trent Gillespie is CEO of Stellis AI and a keynote speaker helping business leaders understand and operationalize AI in their companies. He spent almost nine years leading global innovation efforts at Amazon before leaving to help other companies build the capabilities they need to compete. Book Trent to speak to your group or book a call to discuss using AI within your business.
Tell us what you thought of today's email. |
Did someone forward this newsletter to you? If you're not already signed up, you can subscribe to AI SPRINT™ for free here.