On a gray Tuesday morning in 2022, the Slack notification hit like a small earthquake. Hundreds of employees at a fast-growing software company opened the same blunt message from their CEO: accept a new AI‑driven way of working… or leave with severance. By the end of the week, almost 80% of staff were gone. Some posted angry threads on X, others shared tearful videos on LinkedIn, accusing him of using “AI” as an excuse for brutal cost-cutting. Recruiters quietly added the company to their mental blacklist.
Two years later, the same CEO is sitting on a stage at a tech conference, smiling into the lights. Revenue has doubled. Profit margins have exploded. He leans into the mic and says: “I was right. They were wrong.”
The room goes very, very quiet.
The CEO who bet the company on AI — and his staff’s refusal
The story starts in a glass-walled office in early 2022, when AI was still more buzzword than daily tool. The CEO, let’s call him Daniel, had been playing with early large language models at night, feeding them support tickets, sales scripts, product specs. Something clicked. He saw a different company in his head: smaller, faster, built around AI copilots instead of bloated teams and endless meetings.
He took that vision to his execs. Half loved it, half thought he’d lost his mind. The board asked for numbers. He gave them a promise: cut 70% of repetitive work, shorten release cycles by half, triple output per employee. Everything would depend on one line in a new policy: AI-first, or out.
The internal all-hands where he announced the new rule still lives in employees’ group chats. Daniel stood on stage and told 400 people that their jobs would “only exist in 18 months” if they learned to work with AI agents, not against them. He didn’t sugarcoat. “If you refuse, we’ll help you land somewhere that still works the old way,” he said.
Many heard a threat, not an offer. Designers muttered about being replaced by prompts. Senior copywriters said they weren’t “button pushers for robots.” One engineering manager walked out mid-presentation. Within days, anonymous posts popped up on Blind calling him “the AI cult leader.”
Then came the ultimatum. Employees had 30 days to complete mandatory AI retraining, adopt internal tools, and sign an agreement that their workflows would be redesigned around automation. Roughly 80% either refused or dropped out midway. Some called it a moral stand against “dehumanizing tech.” Others quietly admitted they just didn’t want to relearn their job from scratch.
Daniel didn’t blink. Legal drafted separation packages, HR booked back-to-back exit calls, laptops piled up at reception. Productivity crashed for a quarter, investors panicked, and journalists sharpened their knives. Yet Daniel repeated the same line to anyone who would listen: **“We’re not doing layoffs. We’re changing species.”**
Two years later: fewer people, more output, and messy lessons
Today, that same company runs with a skeleton crew: roughly 90 employees supported by a dense stack of AI agents, scripts, and internal tools. Every customer support interaction is pre-triaged by a model. Every sales deck starts as an AI draft. Code review happens with an AI assistant ghosting alongside human engineers. People aren’t replaced, they are amplified.
➡️ Humanity has just received a 10-second signal dating back 13 billion years
➡️ A Greenland Glacier Is Cracking Open – and Scientists Are Watching It Drain in Real Time
➡️ Retirement: here’s the ideal pension amount for someone living alone
➡️ A gardener reveals why orange peels work wonders in the garden
➡️ No more foil behind the radiators: this far smarter trick warms a room much faster
➡️ Kate Middleton: Her brother James, whose mother is French, celebrates his son’s second birthday (his name is Basque)
➡️ Goodbye to Retiring at 67 Political Storm UK Government Confirms New State Pension Age
➡️ Rental property investing explained: why experts say buying with a mortgage often outperforms paying cash in the long run
The numbers tell a harsh story. Revenue per employee has jumped by more than 300%. Operating margin, once stuck in startup mode, looks like a mature SaaS giant. The company ships new features weekly instead of quarterly. The board members who almost fired Daniel are now parading him in front of other portfolio CEOs as the example of “courageous transformation.”
Walk through their office now and it feels more like a control room than a classic startup floor. Monitors show dashboards of AI agents: success rates, escalations, weird edge cases. A support lead can handle what used to be a 20‑person team, because the model handles 85% of tickets end‑to‑end. A content marketer launches five campaigns a week, because they’re orchestrating prompts, not writing from scratch all day.
One telling metric: the average employee now runs more than 40 automated workflows per day. Two years ago, that number was two. The people who stayed are… oddly calm. Many say they’ll never work in a non‑AI‑native company again, the same way remote workers swear they’ll never go back to five days in the office.
So was the CEO “right”? On the spreadsheets, yes. On the human side, it’s more complicated. The staff who left didn’t all crash and burn. Some joined more traditional firms and are still doing solid, human‑heavy work they love. Others eventually adopted AI on their own terms and admit they might have stayed if the rollout hadn’t felt so brutal.
Daniel’s stance hasn’t softened. He says the refusal wasn’t about ethics, it was about comfort. In his words: **“AI didn’t fire 80% of my staff. Their refusal to adapt did.”** It’s a harsh framing, but it lands because so many leaders now feel the same tension. They see AI’s upside, and the resistance sitting quietly inside their own teams.
What this CEO did that most leaders are still afraid to do
Behind the drama, the playbook he used is surprisingly methodical. First, every role was broken into tasks and patterns. Not titles, not job descriptions — actual day‑to‑day actions. Answering tickets, drafting specs, testing features, cleaning data. Then a small “AI core” team ran experiments: which tasks could be automated, which needed a human in the loop, which were sacred and should stay human-only for now.
Only after that did they roll new workflows out. Each team got a clear picture: here’s what your day looked like before; here’s what it will look like with AI; here’s what that means for your impact and salary. The message wasn’t “AI is coming for your job,” it was “AI is coming for 60% of your tasks — let’s redesign the job around what’s left.”
Most companies skip that messy middle and jump straight from hype to tools. They toss an AI chatbot into Slack, send a link to a webinar, and wait for magic. Then nothing changes. People keep their habits, managers still ask for the same reports, and AI becomes a nice‑to‑have shortcut, not a new operating system.
Let’s be honest: nobody really does this every single day. No one wakes up thinking, “How can I strip my job down and hand half of it to a machine?” That’s why Daniel forced the question. His bet was simple: people who won’t touch AI now will be even more resistant in three years, when the gap is bigger, the fear louder, the stakes higher.
In a podcast last month, he said, “The painful part wasn’t the AI. The painful part was admitting who actually wanted to grow and who just wanted the old days to last a bit longer.”
- He tied AI to compensation: bonuses and promotions were linked to measurable AI leverage, not hours worked.
- He killed legacy reports that AI could auto‑generate, so managers couldn’t quietly cling to old routines.
- He hired differently: curiosity, prompt-writing skill, and comfort with experimentation became non‑negotiables.
- He normalized failure: early AI rollouts were messy, wrong, sometimes embarrassing, and that was allowed.
- He communicated bluntly: *“Your job is changing either with us… or without us.”*
What this story says about us, not just about AI
Underneath the hype and the hot takes, this whole saga is less about technology and more about identity. When a CEO says “we’re going AI‑first,” people don’t hear “new tools.” They hear “the way you’ve created value for 10 years is now negotiable.” That stings. It hits pride, stability, and a quiet fear that the next version of the company won’t need you at all.
We’ve all been there, that moment when a new tool makes you feel like a beginner again in a room where you used to be the expert. Some people lean in and play; others pull back and criticize the tool instead of facing that feeling.
Daniel’s story is extreme, almost brutal in its clarity, and yet it points at a broader shift. Every industry is going to see its version of that Slack message — maybe softer, maybe slower, but the message will rhyme: adapt around AI, or watch your role shrink. Some leaders will choose empathy and long transitions. Others will do what he did: cut fast, rebuild from the ground up.
Neither path is clean. One risks resentment and toxicity. The other risks paralysis dressed up as “protecting people.” The uncomfortable truth is that both sides will tell themselves they were right, using whatever numbers or anecdotes fit their choice.
The question that lingers isn’t just “Was this CEO right?” It’s: what would you do if you got that message today? Would you walk, on principle, to a slower, safer place? Would you stay and wrestle the tools until they become your unfair advantage?
And if you’re a leader, how far would you go to drag your company into the next era — and how many people would you be willing to lose on the way? The answers are going to shape more than earnings calls. They’ll decide who feels at home in the future of work, and who’s left telling stories about how things used to be.
| Key point | Detail | Value for the reader |
|---|---|---|
| AI can radically change productivity | One company doubled revenue with 80% fewer staff by redesigning roles around AI agents | Helps you see what’s realistically possible when AI is treated as infrastructure, not a toy |
| Resistance is mostly about identity, not tools | Many employees refused retraining because it threatened their sense of expertise and stability | Gives language to your own hesitation and to the pushback you may see in your team |
| Leaders need a real playbook, not slogans | Task-level analysis, incentives, and blunt communication mattered more than generic “AI transformation” talk | Offers a concrete pattern to copy or adapt if you’re trying to bring AI into your daily work |
FAQ:
- Question 1Did this CEO really fire 80% of staff just because they refused AI?
- Answer 1According to his own account and multiple employees’ posts at the time, the layoffs were tied directly to a new AI‑first policy and mandatory retraining. People who refused to adopt the new workflows or didn’t complete the program were offered severance and let go.
- Question 2Could the company have reached similar results without such drastic cuts?
- Answer 2Possibly, but it would almost certainly have taken longer. Gradual rollouts tend to preserve more jobs but often dilute the transformation. His approach traded short‑term pain for speed, which is exactly what many investors now praise and many ex‑employees still resent.
- Question 3What does this mean for my own job if I’m not an engineer?
- Answer 3It means your role will likely be unbundled into tasks, and some of those tasks will be handed to AI. Whether you lose power or gain it depends on how quickly you learn to direct, supervise, and remix what the tools produce.
- Question 4As a manager, how can I push AI adoption without breaking trust?
- Answer 4Start by mapping tasks, running small experiments, and being very transparent about what might change. Tie rewards to effective AI use, not just raw hours, and create space for people to admit what they don’t know without feeling stupid or disposable.
- Question 5Is “refusing AI” a realistic stance in the long term?
- Answer 5You can avoid certain tools or industries for a while, but AI is already baked into search, office suites, design tools, and CRMs. The more realistic stance is choosing where and how you engage with it, rather than pretending it won’t touch your work at all.