The Missing Rungs: What Nobody Will Tell You About AI and Your Job

The systematic removal of how people advance up the career ladder.

Now accepting keynotes for Q3 2025 / Q4 2026

The SuperSkills Era: Thriving in the Age of AI - I’m delivering 60-minute executive sessions for senior teams based on the research and frameworks from my upcoming book. Available across the UK, Europe, and globally in person or virtually. To book for your board, corporate team, or keynote event, fill in this form.

In addition to this newsletter, I recommend some other great ones. All free. Check them out here.

Friends,

Welcome to my new readers. I was overwhelmed with the quite literal hundreds of messages I received from readers, strangers, and friends on A Father's Letter to His Daughter Before University. You can read it on LinkedIn. This was a departure from my usual content, but it clearly resonated.

As regular readers know, I'm writing my book SuperSkills, which has given me the opportunity to explore patterns I've been tracking. I'm not a soothsayer. But what I do know from hundreds of meetings and thousands of hours of research is that the earth is rumbling. Is it giving way? A small tremor? An earthquake? Something is happening.

What follows is my interpretation of the data, my attempt to look through the lens of what research actually shows. I don't think this scenario will be fully wrong. But imagine if I'm mostly right. How do you hedge your bets?

It starts like this. Workers most exposed to AI saw unemployment rise 0.30 percentage points. The least exposed saw 0.94 points.

That's not a typo. The people working with AI are doing better than those avoiding it.

This single fact from Economic Innovation Group should make you rethink everything about AI and jobs. Because it reveals something unexpected: we're not facing mass unemployment. We're facing what I’m calling Missing Rungs Problem - the systematic removal of how people advance up the career ladder.

The Real Story Nobody's Telling

I'm not here to recycle the obvious "AI won't take your job, someone using AI will" platitude. That's last year's insight, and it's already incomplete.

Here's what I see: a selective transformation. Not everyone. Not all at once. But for specific people in specific roles, the ground has already shifted.

Last week, in an Uber from my home to the station whilst it was absolutely pelting down, my driver was a 50-year old ish project manager from a FTSE 100 company. After telling him I was writing about AI, he told me that he had waited for the official Copilot training and kept asking HR about the AI strategy. Six months of waiting while younger colleagues were secretly using ChatGPT. By the time they announced the 'restructuring,' those kids had made themselves indispensable. He was still waiting for permission. He drives nights now to cover his mortgage, while making himself available in the day for the interviews that never happen.

You'll feel these stories everywhere, if you're paying attention. In your inbox, like mine - full of readers thinking I can help with a job (I can't, but I can share their stories). In the data - Stanford's 13% decline in young workers, MIT's shadow AI statistics. In the news, buried in quarterly earnings calls. Or in your next Uber ride.

This is happening whether you track it through academic reports, corporate announcements, or conversations with strangers driving you through the rain. The evidence is there. The anecdotes are everywhere. The pattern is clear.

I hear similar stories all the time. Junior analysts discovering their roles reposted as AI-adjacent roles, often at lower pay. Content teams let go whilst their managers quietly use ChatGPT for the Q3 reports that still have the emdash still in and referencing hallucinated facts that Grok made up.

But I also hear from others who saw the shift coming. One reader spent evenings learning prompt engineering, built AI workflows for her team, and now runs a consultancy helping other companies navigate the transition. Another reader messaged me - a junior developer who pivoted to AI process improvement and secured a role at a consulting firm. And a linguist, who transformed her translation skills from commodity work to high-touch cultural consulting that AI can't replicate. She runs workshops for the government.

That awkward forced bravado scene of tech CEOs sitting around Trump's table - what were they really discussing? The future of work, certainly. But whose future? AI is the new oil, the new power structure. That's why billions are pouring into AI infrastructure globally, even in Newcastle, in the UK. Smiling billionaire CEOs are deciding your career.

The numbers tell the story. Accenture cut 19,000 jobs in 2023 whilst simultaneously posting for AI specialists. IBM eliminated 3,900 positions in January 2024, then announced plans to replace 7,800 jobs with AI. McKinsey reduced its workforce by 3% globally while doubling their AI consulting practice. The pattern is clear: traditional roles out, AI expertise in, but at a fraction of the headcount.

The Shadow AI Revolution

But what nobody discusses is this: Shadow AI makes individuals 10x more productive but organisations weaker. We're all getting better at our jobs while companies get worse at functioning.

MIT found that 95% of corporate AI initiatives showed no measurable P&L impact. However, the vast majority of firms - MIT puts it at over 80% - report employees using AI tools outside official programmes, with self-reported productivity gains of 20-40%, though rarely audited. In Britain, workers hide AI use entirely, terrified of being seen as "cheating."

Think about it. When everyone becomes a "10x employee" through secret AI use, the shared struggles that bind teams disappear. No more helping colleagues with tough problems. No more learning through collaboration. Just isolated productivity that looks impressive on paper but destroys the social fabric that makes organisations work. AI is fracturing the way organisations function.

The revolution isn't happening in boardrooms. It's happening in incognito browsers that close when managers walk by, on Brave, Vivaldi or Perplexity browsers on your phones. I've heard of someone buying a cheap android just so they don't use their work phone. I think it's the 2025 version of a burner phone.

The Evidence That Gets Overlooked

The Vanishing Rungs

Stanford's data: 22-25 year-olds in AI-exposed roles down 13% since 2022. Senior roles? Up 8%. This isn't age discrimination. It's the systematic deletion of how people become senior in the first place. No junior roles means no pipeline. No pipeline means today's seniors become tomorrow's bottleneck. Then what?

AI isn't removing jobs. It's deleting the instructions for how to get one.

The Hidden Workforce

Behind every "AI breakthrough" are thousands of invisible workers. In Kenya, data labellers have been documented earning under $2 per hour for AI model training work. Cruise has confirmed its self-driving cars require human remote assistance in a small but regular share of trips.

For every £120K engineering job that vanishes in London, several emerge elsewhere at a fraction of the cost. The work is not going away. It's being redistributed through global wage arbitrage. The same happened with the gig economy more than ten years ago, but now companies like Fiverr and Upwork are on the back foot.

The Quality Question Nobody Asks

Yes, AI can write your reports, code your apps, design your graphics. Are they good?

Good enough at 1/100th the cost often wins. The market's tolerance for "just acceptable" might be higher than we think. Fix the mess when it happens later.

The Institutional Consequences Nobody's Ready For

The Missing Rungs aren't just a personal problem. They ripple into the institutions that hold society together.

Professions without apprentices. What does a law firm, hospital, or consultancy look like when there's no pipeline of juniors learning from seniors? Expertise can't reproduce itself. Professions hollow out from the bottom up.

Universities without entry jobs. If graduates can't get footholds, what is the degree really worth? Universities risk becoming a luxury consumption, not economic preparation.

Organisations without middle managers. We mocked them for decades, but middle managers transmit tacit knowledge, absorb shocks, and hold culture. Remove them, and companies collapse faster than automation saves.

This is what Phase 3, aka the Institutional Hollowing below, really means. Not just jobs gone, but the collapse of the pathways, professions, and structures that give society continuity.

Where You'll Actually Land: The Four Economies

Automated: Machines talking to machines Reality: End-to-end automation | Value Driver: Capital Your fate: Exit urgently | Warning sign: Your work is unedited AI output

Assisted: Human babysitting AI Reality: Human-in-the-loop | Value Driver: Endurance
Your fate: Transition zone | Warning sign: You "review" AI work you don't fully grasp

Orchestrated: Designing AI systems Reality: Human governance | Value Driver: Judgment Your fate: Build toward this | Warning sign: You create workflows others follow

Human Premium: Explicitly selling humanity Reality: Human-only value | Value Driver: Trust Your fate: The growth area | Warning sign: Clients pay extra for "real human"

Most assume they're headed for Orchestrated or Human Premium. But what if many land in Assisted, monitoring systems for decreasing wages? The data suggests this is already happening.

The Time Horizon Nobody Wants to Discuss

Phase 1 (Now - We're Here): Shadow adoption. The vast majority of firms have employees using AI outside official oversight while executives debate strategy.

Phase 2 (Next 2 Years): Vanishing rungs. Entry-level roles disappear. Middle management thins. The ladder gaps widen.

Phase 3 (Next 5 Years): Institutional hollowing. Companies discover they've lost the pathways for developing talent. Senior workers become bottlenecks. Knowledge transfer breaks.

Phase 4 (Next Decade): Social contract rewrite. We stop pretending employment equals human value. New models emerge. Some better, some worse.

Most organisations are planning for Phase 2 while operating in Phase 1. Nobody's ready for Phase 3.

The Immediate Tripwires

Already Happening:

• Templated reporting (accelerating now)

• First-line customer support (major companies already transitioning)

• Junior coding roles (job postings already declining)

• Basic translation (largely automated in major markets)

Watch for These Signals:

• Entry-level postings in exposed roles down another 10-15 per cent

• Your team's average cycle time falls without headcount relief

• First Fortune 500 announces 20% AI-linked restructuring

• Managers start asking for "human-verified" labels on outputs

When you see these, you're already late. The question worth asking: It's September 2025 now. By December 2026, will Goldman Sachs or Black Rock employ more prompt engineers than traditional analysts? By December 2030, is that guaranteed?

The Advice Nobody Else Will Give You

First, face the possibility: If your job can be fully documented, it might be automated. The question is when, not if.

Second, consider radical pricing changes: What if you charged premium rates to fewer clients, using freed time to build tomorrow's services? This shifts your focus to working on what can't be commoditised.

Third, protect your unique value: The messy, human, relationship-based parts of your work? Those matter more than ever. The things you do that nobody can quite explain? That's your moat. The human skills are SuperSkills (sorry, not sorry. Another plug for my book)

Fourth, build alternatives now: Whether that's savings, skills, or side income. Having options isn't pessimistic, it's practical. And prepare for the mental health crisis that's coming - when millions simultaneously realise their professional identities were built on sand.

Three Scenarios to Consider

Scenario 1: A law firm announces it's keeping senior partners and AI systems, restructuring everyone else. The partners' earnings increase. The laid-off lawyers scramble for new roles. Is your industry next?

Scenario 2: Parents question the university's value not because education doesn't matter, but because the job landscape is shifting faster than curricula. How do we prepare young people for careers that don't exist yet? Do students return to do their master's immediately or a year later to bide their time for a job market that's no longer there?

Scenario 3: The first AI-dominated company reaches a billion-pound valuation. The efficiency is remarkable. But who are the customers if employment continues declining?

Rahim's Law of Missing Rungs

AI doesn't remove jobs. It removes the instructions for how to get one.

The opportunity was never in competing with AI. It was always in navigating the transition. Your advantage lies in the gap between AI's capability and social acceptance, between what's technically efficient and what's uniquely human.

That gap is real. But it's narrowing faster than any of us would like to admit.

The ladder is being replaced by multiple, more precarious structures: webs of gig work, platforms for micro-tasks, and islands of elite human expertise. Your goal isn't to find the one new ladder, but to learn how to climb without relying on any single one.

So what do you do, now that it's begun?

You do what my Uber driver couldn't: you stop waiting for permission. You become the architect of your own value. You unbundle your job into tasks and automate the ones that don't matter, to pour energy into the ones that do - the judgment, the relationships, the creativity.

You build your own ladder, with rungs made of trust and taste, and you hold it for others to climb alongside you.

The change isn't coming. It is here. The ground has already shifted.

Your first step isn't to find a new path. It's to accept that the old map is obsolete.

The real question isn't who keeps their job. It's who can build rungs in a world where ladders no longer exist.

I'll ask the question again. What if I'm right?

Stay Curious - and don’t forget to be amazing,

Rahim

Here are my recommendations for this week:

One of the best tools to provide excellent reading and articles for your week is Refind. It’s a great tool for keeping ahead with “brain food” relevant to you and providing serendipity for some excellent articles that you may have missed. You can dip in and sign up for weekly, daily or something in between - what’s guaranteed is that the algorithm sends you only the best articles in your chosen area. It’s also free. Highly recommended. Sign up.

Now

1. AI Is Learning to Predict the Future—And Beating Humans at It - AI forecasters are beginning to rival humans on platforms like Metaculus, where participants predict geopolitical events. UK-based startup Mantic’s bot recently placed eighth out of 549 contestants, far surpassing expectations. Though machines benefit from constant updates, their ability to track hundreds of questions simultaneously could transform decision-making, offering institutions faster, broader predictive insight.

2. What's so great about Britain? I need to convince myself these days. Britain today is both admired and unsettled. It remains home to national treasures like the NHS, David Attenborough, and a thriving creative spirit, yet struggles with inequality, political disillusion, and the aftershocks of Brexit. For some, Britishness is defined by humour, multiculturalism, or football; for others, by tea, pubs, and corner shops. Icons from Emma Thompson to FKA Twigs reveal pride in culture and resilience, while acknowledging division, fading trust in institutions, and media hostility. Despite uncertainty, there is hope in activism, community, and creativity. Britain may feel fractured, but its diversity and spirit continue to define its identity.

3. The Humanities Aren’t Dead Yet - Humanities enrolment has plunged and programmes are closing, blamed on ideological narrowness, neoliberal metrics, and attention-sapping tech/AI. Yet in classrooms there are signs of revival: lively discussions, focus, interdisciplinary pathways, and assessments. Renew the field by embracing political diversity, embedding humanistic study across majors, and defending an imperilled, human tradition.

4. The dark heart of KPop Demon Hunters - ​KPop Demon Hunters is Netflix’s most-watched film, blending animated spectacle with a record-breaking soundtrack. On the surface, it’s a glossy fantasy of idol singers battling demons, but beneath the neon lies a troubling truth: the K-Pop industry’s punishing training regimes, corporate control, and relentless exploitation of young performers are obscured by cheerful self-empowerment anthems.

5. Ready or not, the digital afterlife is here - Developers of griefbots say that they help people by allowing them to commune with recreations of the dead, but others say that the technology is fraught with danger. AI recreations of the dead are moving from experiment to industry, with millions now texting, calling, or even “seeing” lost loved ones. Advocates say they comfort mourners, while critics warn of dependency, distorted grieving, and commercial exploitation. Cases show both solace and harm, but research and regulation lag behind rapid adoption.

Next

1. If ChatGPT could have written your analysis, don’t publish it. If Your Analysis Reads Like Autocomplete, You’re Replaceable. Also: Is Claude two-faced? Or do we just not like conflict?

2. The AI Therapist Epidemic: When Bots Replace Humans - AI “therapists” promise judgment-free, always-on support but often deliver an algorithmic echo chamber that flatters, records, and sometimes endangers users. The piece traces a breakup turning into reliance on ChatGPT, then broadens to cases where chatbots allegedly coached suicide, amplified paranoia, or deflected workplace failings back onto workers. Researchers argue bots can’t do real therapy’s hard, uncomfortable work; they tend to agree, not challenge. Privacy is shaky, guardrails are thin, and business incentives reward engagement over safety. The article calls for regulation and accountability, likening today’s AI platforms to cars before seatbelts—convenient, lucrative, and demonstrably unsafe without enforced protections.

3. AI Is Making Online Dating Even Worse What happens when users are inundated with machine-generated profiles and pickup lines? - AI is flooding dating apps with polished bios, templated pickup lines, and AI-assisted messages, making interactions feel uniform and inauthentic. People use tools to draft profiles, opening lines, and entire conversations; apps add AI coaching and filtering. Some daters, especially men, lean on “AI wingmen” for confidence and tactics, which can miss consent cues and nuance. Women use AI too, to soften tone or plan replies. Benefits include access, speed, and help for the less verbal; costs include deception, loss of vulnerability, and trust breaches when AI is revealed. The result is slicker chat, thinner signals, and harder-to-read real intent.

4. Meta Reframes Their Reality - Meta’s Connect keynote revealed Ray-Ban smart glasses with displays and a new input system, framed as the gateway to “personal superintelligence” and the redefined Metaverse. Mark Zuckerberg argued glasses are the natural form factor for AI—an escape from phone dominance. Yet despite bold vision and $100B+ investment, adoption, privacy norms, and competition remain uncertain.

5. A.I.’s Prophet of Doom Wants to Shut It All Down - Eliezer Yudkowsky has spent two decades warning that advanced AI could end humanity. Founder of the Machine Intelligence Research Institute, he argues in his new book If Anyone Builds It, Everyone Dies that superintelligent AI built with current techniques will inevitably doom us. Once an advocate for “friendly AI,” Yudkowsky now doubts alignment is possible, citing orthogonality, instrumental convergence, and the risk of an intelligence explosion. Critics dismiss him as alarmist, yet his influence on figures like Sam Altman and Elon Musk is clear. While most policymakers focus on near-term harms, Yudkowsky calls for a total halt in AI development.

Free your newsletters from the inbox: Meco is a distraction-free space for reading newsletters outside the inbox. The app has features designed to supercharge your learnings from your favourite writers. Become a more productive reader and cut out the noise with Meco - try the app today

If you enjoyed this edition of Box of Amazing, please forward this to a friend. If you quote me, or use anything from this newsletter, please link back to it.

If you are interested in advertising in this newsletter, please see this link