Artificial Intelligence Content Ideas for Economics
AI's Economic Disruption: A 2028 Crisis Scenario
Featured Image
This scenario models a future where AI's rapid advancement leads to widespread white-collar job displacement, a collapse in consumer spending, and systemic financial risks. The core insight is that AI's ability to substitute for human intelligence, unlike previous technological shifts, creates a negative feedback loop that structurally impairs the economy's earnings and spending power, leading to a potential crisis.
group
Investors Economists
schedule
25 min 0s
open_in_new
citriniresearch.com — AI's Economic Disruption: A 2028 Crisis Scenario
lightbulb Key Insights
Key Insights from Artificial Intelligence Content
1
AI-driven white-collar job displacement can lead to a 'Ghost GDP' where output doesn't circulate through the real economy.
2
The erosion of intermediation layers by AI agents disrupts traditional business models reliant on human friction and brand loyalty.
3
Unlike historical innovations, AI's continuous improvement and cost reduction create a non-cyclical feedback loop of job loss and reduced spending.
4
The white-collar economy's collapse threatens the $13 trillion U.S. residential mortgage market due to impaired income assumptions.
edit_note Content Ideas
Suggestions for topic Artificial Intelligence
Ready-to-use angles — mapped to each distribution channel, with a draft preview.
Twitter/XActionable
Write a 7-tweet thread using the '$180k product manager replaced by an AI agent for $200/month' stat as the hook, then walk through the Human Intelligence Displacement Spiral — one feedback loop per tweet. End with the question: "Which white-collar role in your industry is most at risk?" Format: 7 tweets. Hook strategy: specific dollar comparison creates instant visceral impact. Engagement mechanic: ask followers to name the role they think disappears first.
A $180,000/year product manager. Replaced by an AI agent for $200/month. Here's the economic spiral that follows when this happens at scale:
A $180,000/year product manager. Replaced by an AI agent for $200/month. Here's the economic spiral that follows when this happens at scale:
1/7
That's not a rounding error. That's a 99.9% reduction in labor cost for one of the most common white-collar roles in America. And it's already happening at scale across product, legal, finance, and consulting. The math is brutal.
2/7
Here's what the cascade looks like. Companies lay off white-collar workers → displaced workers take lower-paying gig jobs → wages compress economy-wide → spending power collapses → consumer-facing businesses see revenue fall → those businesses cut more workers to survive → repeat.
3/7
This isn't a normal recession cycle. Normal recessions self-correct: lower rates stimulate hiring, reduced spending leads to restocking. This cycle's engine is AI getting better and cheaper every quarter. Falling demand doesn't slow AI buildout. It accelerates it. Companies substitute AI for headcount to cut costs into a weakening revenue environment.
4/7
The term for what this produces is "Ghost GDP." Output appears in national accounts — corporate profits up, productivity surging, S&P near 8000. But that output doesn't circulate through the real economy. A GPU cluster replacing 500 workers doesn't buy groceries, pay rent, or book vacations. Consumer spending is 70% of US GDP. Machines don't participate.
5/7
The financial system is not priced for this. Private credit grew to $2.5 trillion largely on the back of PE-backed software deals underwritten to perpetual ARR. When AI commoditizes the software those companies sell, ARR stops being recurring. $18B in downgrades in a single month (April 2027) was just the opening act.
6/7
And then the mortgage market. The $13 trillion US residential mortgage market assumes stable borrower income. The prime borrower — 780+ FICO, big down payment, clean history — is a white-collar professional. When that income assumption breaks, the mortgage market doesn't need speculative excess to crack. It just needs a sustained shift in income expectations.
7/7
Which white-collar role in your industry is most at risk right now? The answer you give today is probably the one that starts this loop in your sector. Drop it in the replies.
LinkedInActionable
Write an 850-word analytical post exploring how 'Ghost GDP' — economic output that doesn't circulate through the real economy — could reshape how we measure prosperity. Open with the S&P 500 hitting 8000 while unemployment is rising as the central paradox. Target: investors, economists, policy-focused professionals. Hook strategy: the contradiction between market highs and economic pain forces re-examination. Engagement mechanic: close by asking readers which economic indicator they trust most in an AI-driven economy.
The S&P 500 hit 8000. The unemployment rate hit 10.2%. Here's how both things are true at the same time — and what it means for every investor:
The S&P 500 hit 8000. The unemployment rate hit 10.2%. Here's how both things are true at the same time — and what it means for every investor:
These two numbers should not coexist. In every prior economic framework, a 10.2% unemployment rate would have destroyed corporate earnings, cratered consumer spending, and taken the equity market down with it. The fact that they coexisted — even briefly — is not a statistical anomaly. It is a structural break.
The concept that explains it is "Ghost GDP."
Ghost GDP describes economic output that appears in national accounts but does not circulate through the real consumer economy. A single GPU cluster replacing five hundred white-collar workers produces output — productivity gains, corporate margin expansion, earnings beats — that registers as economic growth. But those five hundred workers no longer earn wages to spend. The output is real. The circulation is not.
This matters because the US economy runs on consumer spending. Roughly 70% of GDP is driven by household consumption. When the entity doing the producing is a machine rather than a person, the output exists but the income velocity does not. GDP can grow while the consumer economy quietly atrophies. Both numbers can be accurate simultaneously. They are just measuring different things.
Investors who anchor to equity multiples and headline GDP miss this divergence entirely.
Here is the sequence that the 2028 scenario models. In 2026, AI-driven layoffs in white-collar roles — product management, software development, financial analysis, legal research, consulting — drove margin expansion. Earnings beat. The market rallied. This is the phase where Ghost GDP begins: output grows, labor income does not.
By 2027, the lagged consumption effect becomes visible. Displaced workers downshift to gig economy roles at a fraction of prior income. A former product manager earning $180,000 takes an Uber driving role at $45,000. This happens across hundreds of thousands of workers simultaneously. The compression is not just at the individual level — it suppresses wages in every sector that absorbs the overflow, because labor supply in those sectors spikes.
Still-employed white-collar workers, watching their peers lose jobs, increase savings rates and pull back on discretionary spending. The behavioral effect precedes the hard data by multiple quarters. High earners use savings buffers to maintain appearances. The consumer economy weakens before the unemployment data confirms it.
The non-cyclical feedback loop is the key structural difference from prior recessions. In a normal downturn, falling aggregate demand eventually slows investment. Companies stop hiring because demand is weak, but they also stop cutting because they've already cut. The cycle finds a floor. Here, falling demand does not slow AI investment — it accelerates it. AI is an OpEx substitution. Companies facing revenue pressure cut headcount and redirect those savings into AI tools to maintain output. The engine driving the job losses does not respond to weak demand signals the way capital expenditure does.
For investors, the practical implications run across four categories.
White-collar consumer discretionary is the most direct exposure. Any company that sells to high-income households — premium retail, travel, financial services aimed at affluent consumers — is exposed to a demand erosion that begins before the unemployment data shows it. The leading indicator is savings rate and credit card utilization among professional cohorts, not headline jobless claims.
Private credit is a second-order risk that is underpriced. The $2.5 trillion private credit market is heavily allocated to PE-backed software and technology deals underwritten on the assumption of perpetual ARR. AI commoditizes software features, collapses differentiation, and enables in-house development at scale. ARR that was assumed to be structurally recurring becomes churn-exposed. The marks on these assets do not yet reflect that. The Zendesk situation — a $5 billion direct lending facility marked to 58 cents after AI agents made its core product obsolete — is a preview, not an outlier.
The mortgage market is the third and most systemic risk. The $13 trillion US residential mortgage book is underwritten on the income assumptions of prime borrowers. Prime borrowers are white-collar professionals. When those income assumptions are structurally impaired rather than cyclically interrupted, the mortgage market faces a risk profile that has no prior template. It is not 2008 — there was no speculative excess in origination. It is something more insidious: a sustained downward revision in the income capacity of the most creditworthy borrowers in the system.
The fourth category is opportunity. The AI infrastructure complex — compute, semiconductor supply chains, energy infrastructure supporting data centers — continues to perform in this scenario because AI buildout is not slowed by weak consumer demand. Economies hosting that infrastructure outperform. Companies that leverage AI as a capability amplifier rather than a cost-cutting mechanism capture value from the transition rather than being consumed by it.
The Ghost GDP concept requires investors to disaggregate what they think they are measuring. GDP growth, earnings growth, and equity multiples were all calibrated for a world where output and income were correlated. That correlation is breaking. The economy can produce more while distributing less. Markets built on the assumption that these move together will be mispriced for longer than feels comfortable.
Which economic indicator do you trust most in an AI-driven economy — and have you updated your framework since 2025?
InstagramActionable
Design a 6-slide carousel titled "The AI Economic Spiral — Explained." Slide 1: hook with the Ghost GDP concept. Slide 2: the feedback loop visualized as a chain (AI → layoffs → less spending → more AI). Slides 3-4: the sectors hit first (SaaS, consulting, payments, real estate). Slide 5: the mortgage market risk. Slide 6: CTA — "Save this to understand what's actually happening in the economy." Hook strategy: turning a complex macro scenario into a visual chain reaction makes it shareable for finance and tech audiences.
The economy grew. Jobs disappeared. Here's the feedback loop nobody is talking about:
Slide 1:
The economy grew. Jobs disappeared. Here's the feedback loop nobody is talking about:
The S&P 500 hit 8000. Corporate profits hit records. Productivity surged. And white-collar unemployment climbed to 10.2%.
These are not contradictions. They are the same machine running at full speed — just not for you.
This is "Ghost GDP." Output that shows up in national accounts but never reaches your wallet. Swipe to see how the spiral works.
Slide 2:
The feedback loop — step by step:
AI gets better and cheaper
→ Companies need fewer white-collar workers
→ Layoffs begin: product managers, analysts, lawyers, consultants
→ Displaced workers take lower-paying gig jobs
→ Wages compress across all service sectors
→ Consumer spending falls
→ Businesses face revenue pressure
→ To cut costs, they buy more AI
→ AI gets better and cheaper
There is no natural brake in this loop. Lower rates do not fix it. The engine runs on AI improvement — and that does not slow down because demand is weak.
Slide 3:
The sectors that got hit first:
SaaS — Agentic coding tools let companies build in-house in weeks. Procurement teams threatened to cancel contracts. ServiceNow announced 15% workforce cuts. Monday.com, Zapier, Asana faced existential pricing pressure.
Consulting — AI agents handled research, analysis, and memo-writing that junior staff used to do. Billing hours collapsed. Firms restructured.
Payments — AI agents bypassed card rails entirely, routing transactions through stablecoins at fractions of a penny. Mastercard's volume growth slowed to +3.4% as agent-led commerce replaced human checkout behavior.
Slide 4:
The sectors that followed:
Real estate — AI agents with MLS access and transaction data replicated what human agents knew. Buy-side commissions compressed from 2.5–3% to under 1%. Volume dropped as white-collar buyers lost income.
Travel and insurance — Agents re-shopped insurance annually, eliminating the 15–20% passive renewal premium. Travel platforms were disintermediated as agents assembled cheaper itineraries in real time.
Financial advice and legal — Routine work automated. Mid-market firms restructured. The long tail of professional services faced the same commoditization that hit SaaS.
Any business model built on human inertia — passive renewals, brand loyalty, friction-dependent pricing — faced structural disruption.
Slide 5:
The risk nobody priced in: the mortgage market.
The $13 trillion US residential mortgage book was underwritten on one core assumption: white-collar borrowers have stable incomes.
780+ FICO scores. Large down payments. Clean credit histories. Prime borrowers.
But "prime" was defined in a world where a product manager's $180,000 salary was durable income. When that income assumption breaks — not from speculation or rate shocks, but from a structural change in the value of human intelligence — the entire book reprices.
Early warning signs in 2027: HELOC draws spiked. 401(k) withdrawals rose. Credit card debt climbed. Mortgage payments stayed current — until they didn't.
The trajectory is the threat. Not the level. Not yet.
Slide 6:
Save this if you want to understand what's actually happening in the economy.
The negative feedback loops described in this scenario had not fully started as of early 2026. That means there is still a window to understand the risks, assess your exposure, and position accordingly.
The canary is still alive. But the mechanism is visible if you know where to look.
Follow for more breakdowns of macro scenarios and structural economic shifts that don't make the front page until it's too late.
YouTube ShortsActionable
Film a 55-second explainer on the "agent-on-agent violence" dynamic — where AI agents negotiating against AI agents destroy the rent-extraction layer built on human limitations (passive subscription renewals, introductory pricing traps, insurance renewal premiums). Open with the DoorDash example: coding agents collapsed the barrier to entry for delivery apps, then AI agents started routing to the cheapest platform in real time. Hook strategy: "agent vs. agent" language creates a vivid conflict frame that drives watch-through.
AI agents are now negotiating against other AI agents. Here's which business models get destroyed first:
[visual cue: split screen — two AI agent interfaces facing each other, price numbers flickering between them]
AI agents are now negotiating against other AI agents. Here's which business models get destroyed first.
[visual cue: DoorDash app logo, then a cascade of competitor app logos appearing around it]
Start with DoorDash. In 2026, agentic coding tools made it possible to build a delivery app in weeks. Dozens of competitors launched. They passed 90 to 95 percent of the delivery fee to drivers to steal market share. Multi-app dashboards let drivers work all platforms simultaneously. Incumbent lock-in evaporated.
[visual cue: consumer AI agent interface, scanning multiple delivery apps in real time, selecting cheapest option]
Then the consumer side. AI agents on your phone checked all delivery platforms for the lowest fee and fastest time — every single order. No loyalty to DoorDash. No loyalty to anyone. The moat of habitual app usage disappeared overnight.
[visual cue: insurance renewal document, agent interface canceling and re-enrolling in real time]
Same pattern in insurance. Agents re-shopped your coverage every renewal cycle. The 15 to 20 percent premium carriers earned from passive human behavior — gone. Real estate commissions compressed from 3 percent to under 1 percent as agents with MLS access replicated what human brokers knew.
[visual cue: credit card transaction being bypassed, stablecoin transfer completing in milliseconds]
And payments. AI agents started routing around card rails entirely — stablecoins on Solana at fractions of a penny per transaction. Mastercard's volume growth slowed to 3.4 percent as agent-led commerce replaced human checkout.
[visual cue: text on screen — "Human inertia was the moat. AI removed the inertia."]
The common thread: every one of these business models was built on human limitations. Time, patience, brand familiarity, passive behavior. AI agents don't have those limitations. When the agent negotiates on your behalf 24 hours a day, the rent extraction layer collapses.
[visual cue: zoom out to show interconnected web of affected industries]
The businesses that survive this are the ones that were already competing on actual value — not on friction.
TikTokActionable
Create a 50-second video contrasting the historical innovation pattern (ATMs → more bank branches, internet → new industries) with AI's structural break: AI improves at the exact tasks humans would redeploy to, eliminating the historical safety valve. Hook strategy: "This time is different" with actual historical data to back it up performs well in economic content on TikTok. Engagement mechanic: ask viewers to comment whether they think new jobs will emerge or not.
Every major technology wave created more jobs than it destroyed. AI is the first one that might not — here's why this time is actually different:
[TEXT OVERLAY: "Every tech wave created more jobs than it destroyed"]
[ACTION: presenter to camera, confident delivery]
Every major technology wave created more jobs than it destroyed. The ATM did not eliminate bank tellers — it made branches cheaper to run, so banks opened more of them. Net result: more teller jobs.
[TEXT OVERLAY: "Internet → 150M+ new jobs globally"]
[ACTION: quick cut to b-roll of tech office, then back to presenter]
The internet destroyed travel agents, video rental stores, print media. And it created e-commerce, social media, cloud computing, digital marketing — entire industries that did not exist before. The pattern held every single time.
[TEXT OVERLAY: "Why did it always work?"]
[ACTION: presenter leans in slightly]
Here is why it worked. Every previous technology automated specific, bounded tasks. It could not do the new jobs that emerged from the disruption. Humans redeployed. The safety valve held.
[TEXT OVERLAY: "AI is different. Here's the actual reason why."]
[ACTION: pause for emphasis]
AI is a general intelligence. It improves at the exact tasks humans would redeploy to. You get displaced from product management — AI is already better at the adjacent roles. You move into prompt engineering — AI gets better at that too, for 200 dollars a month.
[TEXT OVERLAY: "The safety valve is broken"]
[ACTION: slow zoom in]
The new jobs AI creates — AI safety research, model fine-tuning, prompt work — pay a fraction of the roles they replace. And AI is actively improving at those jobs too. The historical pattern required a gap between what machines could do and what humans could do. That gap is closing faster than institutions can adapt.
[TEXT OVERLAY: "Do you think new jobs will emerge this time? Comment below."]
[ACTION: direct eye contact with camera, genuine curiosity in delivery]
This is the first technology wave where that question does not have an obvious answer. Tell me what you think in the comments — new jobs emerge, or this time is actually different?
NewsletterActionable
Write a 900-word investor-focused briefing titled "The Intelligence Premium Unwind: What the 2028 Scenario Means for Your Portfolio." Cover the four systemic risks: labor displacement spiral, Ghost GDP, private credit defaults in PE-backed software, and prime mortgage stress. For each, identify which assets are most exposed and which might benefit. Close with the 'canary is still alive' framing — the negative feedback loops haven't started yet as of early 2026, but the window to position is open. Hook strategy: portfolio action angle converts macro content into practical reader value.
The negative feedback loops haven't started yet. Here's how to position your portfolio before they do — using the 2028 crisis scenario as a map:
The negative feedback loops haven't started yet. Here's how to position your portfolio before they do — using the 2028 crisis scenario as a map:
As of early 2026, the S&P 500 is near all-time highs. The canary is still alive. That is exactly why this is the moment to work through the scenario — not after the feedback loops become self-sustaining, but now, while the positioning window is open.
The 2028 Global Intelligence Crisis scenario, modeled by CitriniResearch, is not a prediction. It is a left-tail risk map. It traces the specific mechanisms through which AI's economic disruption could cascade from sector-specific disruption into systemic crisis. Four structural vulnerabilities are identified. Each has portfolio implications that are actionable today.
**Risk One: The Labor Displacement Spiral**
The scenario's core engine is the Human Intelligence Displacement Spiral. AI capabilities improve → companies need fewer white-collar workers → layoffs increase → displaced workers spend less → margin pressure pushes firms to invest more in AI → capabilities improve further. The distinguishing feature is non-cyclicality. In normal recessions, falling demand eventually slows investment. Here, falling demand accelerates AI adoption — it is an OpEx substitution, not a capex cycle.
Assets most exposed: companies with high revenue concentration in white-collar consumer discretionary spending. Premium retail, travel platforms, luxury goods, financial services targeting affluent professionals. The signal to watch is not headline unemployment — it is the composition of jobless claims (white-collar versus blue-collar) and the savings rate among professional cohorts. Behavioral softening precedes hard data by multiple quarters.
Assets that may benefit: AI infrastructure — compute, semiconductor supply chains, energy infrastructure supporting data centers. The AI buildout does not slow because consumer demand softens. NVDA, TSM, and the energy complex supporting hyperscaler expansion are the direct beneficiaries of a world where every company is substituting AI for headcount.
**Risk Two: Ghost GDP**
Ghost GDP describes economic output that appears in national accounts but does not circulate through the real consumer economy. A GPU cluster replacing five hundred workers produces productivity gains and margin expansion that shows up as GDP growth — but those workers no longer earn wages to spend. Consumer spending is 70% of US GDP. The disconnect between paper wealth and real economy health is the structural feature of this scenario that makes traditional valuation metrics unreliable.
The practical implication: equity multiples calibrated to GDP growth and earnings growth are mispriced if those metrics increasingly reflect Ghost GDP rather than real income circulation. The divergence shows up first in velocity-of-money data and in sector-level revenue growth for consumer-facing businesses. An index-level view masks it. Sector rotation out of consumer discretionary and into capital-light AI infrastructure is the broad directional trade.
**Risk Three: Private Credit Defaults in PE-Backed Software**
Private credit grew from under $1 trillion in 2015 to over $2.5 trillion by 2026. A substantial portion is deployed into PE-backed software and technology companies underwritten on the assumption of perpetual ARR. The scenario's Zendesk case study illustrates the mechanism: a $5 billion direct lending facility marked to 58 cents after AI agents commoditized the company's core customer service product. ARR assumed to be structurally recurring became churn-exposed when AI eliminated the problem the software was solving.
The marks on private credit portfolios do not yet reflect this risk. Public SaaS comparables trade at 5–8x EBITDA; private marks still reflect acquisition multiples from 2021–2022 deals. The lag between mark-to-market and economic reality is a feature of private credit structure, not a bug — until the defaults begin.
The second-order risk is the insurance channel. Alternative asset managers acquired life insurance companies — Apollo/Athene, Brookfield/American Equity, KKR/Global Atlantic — specifically to use annuity deposits as permanent capital for private credit. When regulators tighten capital treatment for privately rated credit held by life insurers, the forced selling dynamic in a seizing market creates contagion risk. Monitor NAIC guidance and RBC factor changes as leading indicators.
**Risk Four: Prime Mortgage Stress**
The $13 trillion US residential mortgage market is underwritten on income assumptions. Prime borrowers — 780+ FICO, large down payments, clean credit history — are predominantly white-collar professionals. When the income-generating capacity of that cohort is structurally impaired rather than cyclically interrupted, the mortgage book faces a risk profile that has no prior template. This is not 2008 speculative excess. It is a sustained downward revision in the income capacity of the most creditworthy borrowers in the system.
Early warning signals in the scenario: HELOC draws, 401(k) withdrawals, and rising credit card debt appearing quarters before mortgage delinquencies. Geographic concentration matters — zip codes with greater than 40% tech and finance employment show stress first (San Francisco down 11% year-over-year, Seattle down 9%, Austin down 8% in the June 2028 Zillow data). Monitoring HELOC utilization rates and early-stage delinquency data by zip code provides lead time before the broad mortgage risk becomes consensus.
**Positioning Framework**
The scenario is not a call to de-risk today. It is a call to audit your exposure against four specific mechanisms that are not yet in the consensus view.
Reduce unexamined exposure to: white-collar consumer discretionary revenue concentration, private credit portfolios with PE-backed software allocation at 2021-era marks, and residential real estate in high-tech-employment zip codes.
Maintain or increase exposure to: AI infrastructure (compute, energy, semiconductors), companies demonstrating AI-as-capability-amplifier rather than AI-as-cost-cutter, and assets that benefit from the continued buildout regardless of consumer demand.
Watch for inflection: the composition of initial jobless claims shifting toward white-collar professionals is the macro signal that the spiral has begun. HELOC draw acceleration in tech-heavy zip codes is the mortgage signal. NAIC capital guidance changes are the private credit signal.
The negative feedback loops have not started yet. The canary is alive. The 2028 scenario is most valuable right now — as a map of what to watch before it becomes obvious.
quiz Frequently Asked Questions
Economics & Artificial Intelligence: Common Questions
Answers to the most common questions about creating Economics content around Artificial Intelligence topics.
This is explicitly modeled as a scenario — a thought exercise designed to prepare readers for potential left-tail risks, not a prediction. As of early 2026, the S&P 500 is near all-time highs and the negative feedback loops described have not yet begun. The value of the scenario is as a risk map: identifying which economic mechanisms (Ghost GDP, private credit exposure, prime mortgage stress) are structurally vulnerable to AI-driven disruption, so investors and businesses can assess their own exposure before the feedback loops start. The 'canary is still alive' — there is time to act.
Ghost GDP describes economic output that appears in national accounts (GDP figures, corporate earnings) but does not circulate through the real consumer economy. A single GPU cluster replacing thousands of white-collar workers produces output that shows up as productivity gains and corporate profit — but those replaced workers no longer earn wages to spend. Since consumer spending drives roughly 70% of US GDP, Ghost GDP is a structural disconnect: the economy looks healthy on paper while the consumer economy quietly atrophies. Investors focused solely on equity multiples and GDP growth miss this divergence.
The spiral describes a self-reinforcing feedback loop: AI capabilities improve → companies need fewer workers → white-collar layoffs increase → displaced workers spend less → margin pressure pushes firms to invest more in AI → AI capabilities improve further. Unlike normal recessions, which have self-correcting mechanisms (lower rates stimulate hiring, reduced spending eventually leads to restocking), this cycle's cause is AI's continuous improvement and cost reduction. Falling aggregate demand does not slow AI buildout — it is an OpEx substitution, meaning companies can cut headcount and buy more AI simultaneously.
The $13 trillion US residential mortgage market is built on the assumption of stable borrower income. When white-collar workers — the prime borrowers with 780+ FICO scores, significant down payments, and clean credit — face sustained income impairment from AI displacement, the income assumptions underlying those mortgages become structurally invalid. Unlike 2008, which was driven by speculative excess, this crisis stems from a fundamental change in the value of human intelligence. Early stress signals include HELOC draws, 401(k) withdrawals, and spiking credit card debt even while mortgage payments remain current — the lagged pattern of people using savings to maintain appearances before the hard data confirms the problem.
AI agents removed the friction that allowed businesses to extract rent from human limitations — time, patience, brand familiarity. Subscriptions that passively renewed were negotiated down. Introductory pricing that doubled after trials was caught and cancelled. Insurance renewals were re-shopped annually. Real estate commissions were compressed from 2.5-3% to under 1% as AI agents with MLS access replicated the knowledge base of human agents. Any business model predicated on human inertia — what the scenario calls "habitual intermediation" — faces structural disruption as AI agents optimize continuously on behalf of consumers.
Historical technological innovation destroyed jobs but created more — ATMs led to more bank branches, the internet created entirely new industries. This pattern held because new jobs required human performance at tasks AI could not do. AI, as a general intelligence, improves at the very tasks humans would redeploy to, breaking the historical safety valve. New AI-created jobs (prompt engineers, AI safety researchers) pay a fraction of the roles they render obsolete. The scenario identifies this as the first time the most productive asset in the economy has produced fewer jobs — not because it eliminates all work, but because it improves faster than humans can adapt.
The scenario identifies four structural vulnerabilities: white-collar labor displacement affecting consumer spending, Ghost GDP divergence between paper wealth and real economy health, private credit defaults in PE-backed software deals with inflated ARR assumptions, and prime mortgage stress in zip codes with high tech and finance employment concentration. Portfolio positioning implications include scrutinizing exposure to companies dependent on white-collar consumer discretionary spending, evaluating private credit exposure to software-dependent businesses, and monitoring leading indicators like HELOC draw rates and initial jobless claims composition (white-collar vs. blue-collar) as early warning signals.
The scenario's practical value is in identifying which business models are structurally exposed to AI disruption before the feedback loops become obvious. Businesses reliant on intermediation (real estate, insurance renewals, SaaS subscription lock-in, financial advice) should stress-test their revenue model against AI agent disintermediation. Businesses that leverage AI as a capability amplifier rather than a cost-cutting tool — the IKEA model of retraining employees as AI-assisted designers — are better positioned to capture value from the transition. The scenario suggests that the window to reposition is 2026, before the feedback loops that accelerated through 2027 become self-sustaining.
map Explore Related Ideas
Related Economics Strategies
rocket_launch
Turn Any Economics URL into Content
Paste any Artificial Intelligence article, video, or podcast into PullContent and get platform-ready drafts, key insights, and content angles in seconds.
add_link Start Mining Your Next Viral Postbuild Explore Free Tools