.svg){kind=logo}
Research & Insights
AI won't kill jobs. It will redistribute them. And that changes everything.
A response to Jack Dorsey's decision to cut nearly half of Square's workforce, and what it actually signals about the future of work.
A response to Jack Dorsey's decision to cut nearly half of Square's workforce, and what it actually signals about the future of work.
Jack Dorsey just announced he’s cutting Block’s workforce from over 10,000 to just under 6,000. Nearly half the company, gone in a single day. His reasoning was direct: AI and smaller, flatter teams are enabling a fundamentally different way of building and running a company, and he’d rather act now than bleed slowly.
The internet, predictably, split into two camps. One side says this is the inevitable future — AI replaces humans, and we should all brace for mass unemployment. The other side says this is corporate greed dressed up in a technology narrative.
I think both are wrong. What’s actually happening is more nuanced, more structural, and ultimately more optimistic than either camp admits — though the path there won’t be painless.
The emerging evidence supports this. Recent analysis from BCG estimates that over the next two to three years, 50% to 55% of US jobs will be reshaped by AI.
Not eliminated. Reshaped.
Most people will keep their roles but face radically different expectations for how they work and what they produce. That distinction matters enormously, and it is the one most commentary ignores.
Let’s be honest about something the tech industry has avoided saying out loud for years: large organizations are bloated. Not because leadership is incompetent, but because scale itself creates inefficiency. When you have 10,000 people, you don’t just have 10,000 workers — you have layers of coordination, approval chains, internal communication overhead, and entire teams that exist primarily to manage the complexity created by having so many teams in the first place.
AI doesn’t just automate individual tasks. It collapses the coordination cost that justified many of those roles. When an AI agent can summarize a document, draft a response, route it to the right stakeholder, and track follow-ups — that changes things. You no longer need the people whose job was to be connective tissue between departments.
This is not speculation. Research from Harvard Business School found that since the public launch of ChatGPT, job postings for structured and repetitive roles — the connective tissue roles — decreased by 13%.
At the same time, demand for analytical, technical, and creative work grew by 20%. The labor market is already repricing the kind of work that large organizations produce the most of.
So yes, Dorsey is right. Large enterprises will compress headcount, and significantly. But here’s the part most people miss: this dynamic is largely confined to organizations that had the slack to absorb it. The fat exists at scale. A 50-person company doesn’t have three layers of middle management. They don’t have a team of six whose primary job is internal reporting. They were already lean.
For small and mid-sized businesses, AI tells a completely different story. These companies were never overstaffed — they were understaffed. They were already doing more with less, often competing against larger players with a fraction of the resources.
AI changes that equation dramatically. A 30-person company with the right AI infrastructure can now produce the output of a 100-person company.
A 200-person company can compete with enterprises ten times their size. The key is multiplying the capability of every person they already have.
This is the part of the AI story that doesn’t get enough attention: AI levels the competitive playing field. Not completely — capital, brand, and distribution still matter — but meaningfully. The gap between what a small team and a large organization can accomplish is narrowing faster than ever before.
The mechanism is augmentation, not replacement. Harvard Business School research describes it clearly: generative AI creates new demand in augmentation-prone roles, where human-AI collaboration becomes the primary driver of output.
For small companies, this is the entire point. Every person on the team becomes more capable, not more expendable.
Headcount is becoming decoupled from output. Revenue per employee will become the defining metric of the next decade. Small, AI-native companies will set records that would have been unthinkable five years ago.
This is where the conversation gets most important, and where I think the prevailing narrative is most wrong.
The assumption embedded in the “AI kills jobs” argument is that when large enterprises lay people off, those people have nowhere to go. That the economy is a fixed pie, and if big companies need fewer humans, there are simply fewer jobs to be had.
History tells us the opposite. Every major technology shift — the printing press, electricity, the internet — initially displaced workers from established institutions. And every time, it simultaneously lowered the barrier to creating new institutions.
And the early data from this shift is consistent with that pattern. Anthropic's labor market research found no systematic increase in unemployment for workers in the most AI-exposed occupations since late 2022.
Displacement is happening. But so is absorption — faster than the headline layoff numbers suggest.
AI is doing this right now, at an unprecedented pace. The cost and complexity of starting a business is collapsing. What once required a team of 20 — engineers, designers, marketers, customer support — can now be built by three or four people with the right AI tools.
Product development cycles that took 18 months can happen in 18 weeks. A solo founder can build, ship, and support a product that would have required venture funding and a full team just five years ago.
This means we’re about to see an explosion of new businesses. Micro-businesses, niche SaaS companies, specialized agencies, AI-native service firms. And those businesses will need experienced people — exactly the kind of people being laid off from large enterprises. People who understand how to operate at scale, how to manage complex products, how to serve demanding customers. That institutional knowledge doesn't disappear. It redistributes.
We can already see this dynamic in the data. A 2026 survey from Epoch AI and Ipsos found that while 20% of full-time workers say AI has taken over parts of their job, 15% have started doing entirely new tasks they would not have done without AI.
Roles are not vanishing into a vacuum. They are being rewritten.
That institutional knowledge does not disappear. It redistributes.
I would be doing you a disservice if I made this sound clean and seamless. It won’t be.
The timing mismatch is real. Layoffs happen in a single announcement. New business creation takes months, sometimes years. In between, there are real people facing real financial pressure, uncertainty, and the psychological toll of an involuntary career disruption. The macro trend being positive doesn’t make the micro experience any less difficult.
Not everyone will become a founder, and they shouldn’t have to. Many of the people leaving large enterprises want — and deserve — stable employment with benefits, predictable income, and career growth. The emerging ecosystem of smaller companies will eventually provide this, but it will take time for that infrastructure to mature.
In the meantime, the burden falls on both companies and policymakers. Harvard Business School research is direct on this point: retraining is essential, especially in roles where AI is reducing skill diversity.
Workers in automation-prone occupations face displacement unless they develop non-automatable skills — judgment, interpersonal communication, contextual decision-making. The transition will not manage itself.
And we should acknowledge that large enterprises won’t just shrink and retreat. They’ll become leaner, faster, and potentially more dominant with fewer people. Dorsey isn’t downsizing because Block is struggling — he’s restructuring to accelerate. The competitive pressure on smaller companies could actually intensify before the leveling effect of AI fully plays out.
Here’s the thesis I want to leave you with.
We’re not heading toward a world with fewer jobs. We’re heading toward a world where employment is distributed differently. For decades, the trajectory has been toward concentration — large enterprises absorbing more and more of the workforce, becoming the default employer for millions. AI reverses that trend.
The future of employment looks less like 50 people at one massive company. It looks more like 50 people across 10 companies — each one leaner, more focused, and more capable than before AI. The total number of jobs may stay roughly the same, or even grow, but they’ll be spread across a much larger number of organizations.
This is, I believe, ultimately a healthier economic structure. More companies means more competition, more innovation, more resilience, and more paths to building wealth. It means fewer people dependent on a single employer’s decisions — like, say, a CEO deciding over a weekend to cut 4,000 jobs.
If you’re running a large organization, the lesson from Dorsey’s announcement isn’t necessarily to follow his exact playbook. But you should be asking hard questions about which roles exist because of genuine value creation and which exist because of organizational complexity. AI won’t just let you do the same things with fewer people — it will let you do entirely different things with a fundamentally different team structure.
But the balance matters. BCG's analysis warns that leaders who cut their workforce beyond AI's current ability to replace it will see productivity drop. The imperative is not maximum reduction.
It is the right balance of automation, upskilling, and deliberate talent planning. Workforce strategy cannot sit downstream of AI adoption. It must be embedded in it.
If you’re running a small or mid-sized company, this is your moment. AI is your equalizer. The companies that will thrive aren’t the ones that use AI to cut costs — they’re the ones that use it to expand what’s possible with the team they already have.
And if you’re one of the 4,000 people leaving Block today — or anyone watching this unfold and wondering about their own future — understand this: the economy is not shrinking. It’s reshaping. Your skills, your experience, your ability to build and execute — those things are about to be in demand in places that didn’t exist two years ago.
It is worth noting who tends to be most exposed. Anthropic's research found that workers in the most AI-affected professions are more likely to be highly educated and higher-paid.
These are not entry-level roles being erased. These are experienced professionals whose institutional knowledge is exactly what the next wave of companies will need.
The jobs aren’t disappearing. They’re moving.
The fear of AI replacing jobs focuses on repetitive, routine work first — not jobs wholesale. The roles most at risk share one trait: their core tasks are structured and predictable enough for a machine to execute.
Categories drawing the most displacement pressure include administrative and clerical roles (data entry, scheduling, bookkeeping), customer service and sales (tier-1 support, telemarketing), finance and legal support (invoice processing, contract review), media and content (templated copywriting, basic design, entry-level coding), and logistics (route optimization, dispatch coordination).
HBS research found that after ChatGPT launched, postings for automation-prone roles dropped 13% while postings for AI-enhanced roles grew 20%. The work is shifting, not simply vanishing.
AI can automate a task. It cannot replace judgment under pressure, physical dexterity in unpredictable environments, or genuine human connection.
Resilient roles include healthcare and mental health (nurses, therapists, doctors, social workers) and skilled trades (electricians, plumbers, HVAC technicians). Strategic leadership and creative direction — requiring original vision and judgment — are also well protected.
The pattern across research is consistent: the least-at-risk jobs require high emotional intelligence, non-routine physical skill, or decisions that carry real-world stakes and accountability.
Not replace — reshape. The narrative of AI replacing jobs wholesale overstates the near-term risk. BCG projects that 50–55% of US jobs will be substantially changed by AI over the next two to three years.
Full job elimination follows a slower curve: roughly 10–15% of roles over the next five years. By 2030, the tasks inside most jobs shift, structured/repetitive roles face the steepest pressure, demand for analytical and interpersonal skills grows, and new roles emerge.
Anthropic's research adds that the most exposed workers tend to be older, more educated, and higher-paid. Risk is not evenly distributed and is not simply a story about low-wage work.
The impact is already here and measurable. A 2026 survey by Epoch AI and Ipsos of 2,000 American adults found key results. Some 20% of full-time workers say AI has replaced parts of their job; 15% say AI created new tasks they wouldn't have had otherwise.
Half of all American adults used AI in the past week. Replacement is currently outpacing augmentation, but both are happening simultaneously.
The companies and workers who will fare best are not waiting to see how it plays out. They are already deciding how to respond.
.svg)
Lucidya is the leading AI-native platform for global customer experience intelligence. With its powerful multilingual sentiment and tone capabilities, our platform is designed to give brands the power to deliver game-changing, deeply personal customer experiences across any market.
Lucidya connects all your customer-facing channels — social, media, surveys, and support — into one intelligent system. It turns raw data into actionable insights so your teams can monitor sentiment,tailor messaging, protect reputation, boost satisfaction, all in real time.
Generic AI simply processes text, but our proprietary, in-house AI is built to understand emotion. By mastering sentiment and tone across a massive range of global languages, we provide the unmatched clarity your teams need to respond with absolute confidence.
Yes. Lucidya complies with Saudi PDPL, GDPR, and SOC2 standards. Data is encrypted, securely stored, and can be hosted regionally to meet compliance needs.
Lucidya is the leading platform for customer experience management in the Arab World. With unique AI and NLU capabilities, this CXM platform is designed to give brands the power to deliver game-changing customer experiences anywhere in the region.
Lucidya is the leading platform for customer experience management in the Arab World. With unique AI and NLU capabilities, this CXM platform is designed to give brands the power to deliver game-changing customer experiences anywhere in the region.
Lucidya is the leading platform for customer experience management in the Arab World. With unique AI and NLU capabilities, this CXM platform is designed to give brands the power to deliver game-changing customer experiences anywhere in the region.
Lucidya is the leading platform for customer experience management in the Arab World. With unique AI and NLU capabilities, this CXM platform is designed to give brands the power to deliver game-changing customer experiences anywhere in the region.
.svg)
.svg)
