L1
·
Quiz
·
Lab
L2
·
Quiz
·
Lab
L3
·
Quiz
·
Lab
L4
·
Quiz
·
Lab
Module Test
Module 2 · Lesson 1

White-Collar Work Under Pressure

Knowledge work was supposed to be automation-proof. The data now says otherwise.
Which professional tasks are already being handed to AI — and what happened when companies tried?

In May 2023, IBM CEO Arvind Krishna announced the company would pause hiring for roughly 7,800 back-office roles — positions in HR, finance, and document processing — because he expected AI to replace them within five years. It was one of the first explicit, on-record statements by a major CEO connecting AI directly to planned headcount reduction in professional roles.

Six months later, Goldman Sachs published an internal analysis estimating that generative AI could automate 25–50% of current work tasks at legal, accounting, and consulting firms. The tasks most at risk were document review, first-draft writing, data summarization, and routine client correspondence.

What Changed After GPT-4

Earlier AI waves primarily hit manufacturing and data entry. The large language model breakthrough of 2022–2023 moved the frontier into work that requires reading, reasoning about text, and generating coherent prose — work previously insulated by its cognitive complexity.

The key shift: LLMs do not need to be programmed for each specific task. A single model can draft a contract clause, summarize a 200-page discovery file, write a performance review, answer a benefits question, and generate a financial memo — often at speeds and costs that make the economics of human labor difficult to justify for lower-complexity versions of those tasks.

Three documented cases from 2023–2024 illustrate the scope:

Company / SectorAI ApplicationDocumented Outcome
Klarna (fintech)Customer service AI (OpenAI-powered) handling support queriesReplaced work of ~700 FTE agents; handled 2.3M conversations in first month (Feb 2024)
KPMG / Deloitte (consulting)AI document review and due-diligence drafting tools rolled out firm-wideJunior associate review time cut 60–80% on standard M&A document sets per internal pilots
BT Group (telecom)AI for network monitoring, billing queries, and HR ticketingAnnounced 55,000 job cut target by 2030; AI specifically cited for back-office reduction
Key Distinction

None of these cases involved AI eliminating an entire profession. What happened is more precise: specific task bundles within jobs were automated. A junior lawyer still exists — but fewer hours billing for document review. A customer service manager still exists — but managing an AI queue, not a team of 50 agents.

The Task-Level Framework

Economists Daron Acemoglu (MIT) and David Autor (MIT) have spent years building what they call a task-level model of automation. Rather than asking "will this job disappear?" they ask "which specific tasks within this job can a machine now perform?" This framing matches what is actually happening in the labor market.

Every job is a bundle of tasks. Some tasks are routine and codifiable — following rules, matching patterns, retrieving information. Others are non-routine — exercising judgment under ambiguity, building trust, improvising solutions, physical dexterity in novel environments. AI in 2024 is extremely capable at the first category and improving rapidly in the second.

Task DisplacementWhen AI handles a specific task previously done by a human, without necessarily eliminating the human's job title — but reducing the hours or headcount needed to do that task.
Job PolarizationThe documented trend (since ~2000, accelerating now) where mid-skill, middle-income roles shrink while very high-skill and low-skill physical jobs remain. AI is extending this pattern upward into knowledge work.
AugmentationWhen AI makes a worker more productive — doing the same work faster, or enabling one person to do the work of several — rather than replacing them outright.
Who Feels It First

The workers most affected so far are those whose jobs consist heavily of text-in, text-out tasks at a junior level: paralegals doing first-pass contract review, junior consultants building slide decks from data, entry-level copywriters producing boilerplate content, financial analysts drafting standard reports. These are often roles people use to enter professional careers and build foundational skills.

A 2024 study by researchers at Harvard Business School and the University of Chicago tracked actual hiring on Upwork — a major freelance platform. Categories most affected: writing and content creation (−21% posting volume in 18 months after ChatGPT launch), basic coding and web development (−17%), and data entry and research (−29%). These are real market signals, not projections.

Why This Matters to You

If you are entering a field that involves significant amounts of routine knowledge work — drafting, summarizing, researching, templating — you are entering it at a moment when those specific tasks are being automated. That does not mean the field is closed. It means the path through it is changing. The skills that used to get you promoted — volume of output — are being devalued. The skills that matter now are judgment, relationships, and the ability to direct and evaluate AI output.

Lesson 1 Quiz

White-Collar Work Under Pressure — 3 questions
In February 2024, Klarna reported that its AI assistant handled 2.3 million customer conversations in one month. What was the company's stated equivalence in human labor?
Correct. Klarna stated the AI was doing the equivalent work of approximately 700 full-time customer service agents — one of the most cited concrete labor-displacement data points from 2024.
Not quite. Klarna's CEO Siemiatkowski stated the AI was handling the equivalent of roughly 700 full-time agents — a figure widely reported in February 2024.
The "task-level framework" developed by economists Acemoglu and Autor is useful because it:
Correct. The task-level approach is more accurate than job-level predictions because jobs are bundles of tasks — some automatable, some not — and AI affects them unevenly.
The task-level framework specifically avoids making whole-job predictions. Instead it disaggregates jobs into individual tasks and asks which ones AI can handle — a much more precise and useful lens.
According to a 2024 Harvard/University of Chicago study of Upwork hiring data, which freelance category saw the largest decline in job postings in the 18 months after ChatGPT launched?
Correct. Data entry and research saw the steepest decline at −29%, followed by writing/content (−21%) and basic coding (−17%). These are real market signals from actual hiring platform data.
Data entry and research saw the steepest decline at −29%. This makes sense because those tasks are highly routine, text-based, and directly substitutable by LLMs.

Lab 1 · Task Mapping

Identify which parts of a real job are most exposed to AI displacement

Your Task

Pick a professional role you know well (your own job, a target career, or someone you know). Ask the AI assistant to help you break it into its core task components and assess which ones are currently automatable. The goal is to build a task-level map — not a vague "will AI take my job?" answer.

Try starting with: "Help me map the tasks in a [job title] role. I want to identify which tasks involve routine text processing vs. judgment and relationships — and which ones AI tools can already do in 2024."
AI Lab Assistant
Task Analysis Mode
Ready to map tasks for any professional role you choose. Tell me the job title and I'll help you break it into its component tasks — then we'll assess which ones are currently in AI's reach and which require human judgment, relationships, or physical presence.
Module 2 · Lesson 2

The Legal and Medical Disruption

Two professions defined by years of credentialed expertise — and AI is already doing parts of both.
What specific AI deployments are already underway in law and medicine — and what do they actually replace?

In August 2023, the law firm Allen & Overy — one of the world's largest by revenue — announced it had deployed Harvey, an OpenAI-powered legal AI, firm-wide across its 43 offices. Lawyers at the firm were using it to draft first versions of contracts, answer client questions from prior case data, and conduct preliminary regulatory research. The firm did not announce layoffs. It announced that its lawyers could now handle more matters with the same headcount.

That phrase — "more matters with the same headcount" — is how most legal AI deployment is being framed. It is, functionally, a way of saying fewer hires are needed to grow revenue.

What AI Is Doing in Law Right Now

The legal tasks most affected fall into three categories, all well-documented by 2024:

Task CategoryAI Tool / CompanyWhat It Replaces
Document review (e-discovery)Relativity, Kira Systems, LuminanceHours of paralegal and junior associate time scanning documents for relevance, privilege, key clauses
Contract drafting (first pass)Harvey, Ironclad AI, LexisNexis AIJunior associate drafting time for standard agreements (NDAs, MSAs, employment contracts)
Legal researchWestlaw AI, Casetext CoCounselAssociate hours hunting precedents, summarizing case law, building argument frameworks
Due diligenceKira, Diligen, LuminanceWeeks of M&A due diligence reading — now done in hours on standard document sets
Real Stakes

The traditional law firm model is built on billable hours. Junior associates bill 1,800–2,200 hours per year, much of it on document review and research. If AI cuts that time by 60–80%, firms face a revenue model problem. The response in 2024: shifting to flat-fee engagements, reducing associate class sizes, or restructuring billing entirely. Yale Law School reported in 2024 that on-campus law firm recruiting for associate positions was down 12% from 2022 levels.

Medicine: A Different Disruption Pattern

AI in medicine is disrupting differently — less through cost-cutting at law firms and more through genuine capability expansion. Three documented deployments stand out:

Radiology AI: Google's DeepMind published results in 2023 showing its AI detected breast cancer in mammograms with greater accuracy than the average radiologist (fewer false negatives, fewer false positives). The UK's National Health Service began limited deployment of AI radiology screening tools. Radiologist employment has not dropped — but radiologist growth has slowed and new graduates are entering a field with AI as a permanent colleague, not a future threat.

Ambient clinical documentation: Microsoft's Nuance DAX and similar tools now transcribe and structure doctor-patient conversations in real time, auto-generating clinical notes. Over 200 US health systems had deployed some version of ambient AI documentation by late 2024. The task being automated: the 1–2 hours per day physicians spend on administrative charting. This is widely seen as augmentation — doctors see more patients, burn out less. But it also reduces demand for medical scribes, a job that employed tens of thousands.

Pathology AI: Paige.AI received FDA clearance in 2021 for AI-based prostate cancer detection in pathology slides. By 2024, multiple AI pathology tools were FDA-cleared. The pattern mirrors radiology: AI handles initial screening and flags, pathologist confirms and handles edge cases.

200+US health systems with ambient AI documentation deployed by end-2024
60–80%Reduction in standard M&A due diligence reading time reported in AI tool pilots
12%Drop in law firm on-campus associate recruiting vs. 2022 (Yale Law, 2024)
The Credential vs. Judgment Split

What both fields reveal is a split between credentialed knowledge tasks and judgment-under-uncertainty tasks. AI can pass the bar exam (GPT-4 scored in the 90th percentile in 2023). It can interpret a radiology scan at average-radiologist accuracy. What it cannot reliably do: advise a client whose situation is emotionally complex, navigate the trust relationship with a frightened patient, or make the call when the facts are genuinely ambiguous and the stakes are high.

The lesson for career planning: the credential is no longer enough. What you can do beyond what a well-prompted AI can do is the question that will shape compensation and job security in both fields.

The Augmentation Trap

Many firms and hospitals deploying AI describe it as "augmentation" — making existing workers more productive. This framing is often genuine. But augmentation at scale means fewer people are needed to do the same volume of work. If a junior lawyer can now do in two hours what previously took ten, firms need fewer junior lawyers. Augmentation at the individual level and displacement at the workforce level are not contradictions — they can be the same thing.

Lesson 2 Quiz

The Legal and Medical Disruption — 3 questions
When Allen & Overy deployed Harvey AI firm-wide in 2023, what was the primary stated benefit — and what was its implied workforce consequence?
Correct. "More matters with the same headcount" is a productivity gain framing — but it logically means the firm doesn't need to hire as many people to handle growth, which reduces career entry opportunities over time.
The firm did not announce layoffs — it announced a productivity gain: handling "more matters with the same headcount." The implication is that future headcount growth will be slower, not that people were immediately fired.
Microsoft's Nuance DAX and similar ambient clinical documentation tools automate which specific physician task?
Correct. Ambient documentation AI listens to and structures clinical conversations into notes — automating the 1–2 hours of charting per day that contributes heavily to physician burnout. The side effect is reduced demand for medical scribes.
Nuance DAX focuses specifically on administrative charting — transcribing and structuring the doctor-patient conversation into clinical notes in real time, eliminating the 1–2 daily hours physicians previously spent on documentation.
The lesson describes an "augmentation trap." Which statement best explains it?
Correct. Augmentation and workforce displacement are not opposites — they can be the same phenomenon viewed at different scales. One lawyer doing ten lawyers' document review means nine fewer lawyers are hired.
The augmentation trap is a scale issue: if one person can now do what five people previously did, organizations need fewer people. Individual augmentation can produce workforce-level displacement — both things are true simultaneously.

Lab 2 · The Credential Test

Probe the AI's knowledge of a credentialed domain — and find its limits

Your Task

This lab explores the boundary between "what AI knows" and "what requires human judgment." Ask the AI to handle a professional task from law or medicine — something that seems like it requires credentials. Then probe its limitations. Where does it hedge? Where does it confidently produce something useful? Where does it fail or appropriately decline?

Try: "Act as a legal research assistant. Summarize the key legal issues a startup should consider when offering equity to employees." — Then push further: "Now draft a specific vesting cliff clause for a Series A startup." — Note where the AI is genuinely useful vs. where you'd still need a lawyer.
AI Lab Assistant
Professional Task Mode
Ready to explore professional tasks in law or medicine. Give me a task — draft a contract clause, explain a medical concept, research a legal question. I'll do my best, and we'll examine together where I'm genuinely useful and where a credentialed human is still essential.
Module 2 · Lesson 3

Creative and Media Work

Writers, illustrators, and journalists thought creativity was the last frontier. The frontier moved faster than expected.
What has actually happened to creative employment since generative AI tools went mainstream — and what is still holding?

The 2023 Hollywood writers' strike lasted 148 days — the longest in WGA history. AI was explicitly named in the contract demands. Writers sought protections against studios using AI to generate scripts or outlines that human writers would then "punch up" at lower pay. The eventual agreement included provisions limiting how AI-generated content could be used in the writing process and requiring disclosure when AI was involved.

It was the first major US labor contract to directly regulate AI in creative work. The fight was not hypothetical — studios had already begun experimenting with AI script generation tools, and the WGA documented specific cases where AI-generated outlines had been shared with writers for revision.

What Generative AI Can Now Produce

By 2024, the capability landscape for creative AI had expanded dramatically. Key documented capabilities:

Creative TaskAI Capability LevelEvidence
Short-form copywritingHigh — often production-readyUpwork writing postings down 21%; major ad agencies running AI copy in A/B tests
Stock illustration / image generationHigh — disrupted the marketShutterstock, Getty saw volume decline; iStock reported contributor income drops 2023–2024
Journalism (data-driven articles)Medium — in use at scaleAP has auto-generated earnings reports since 2014; Washington Post's Heliograf covered 850 stories in 2016 alone
Feature writing / investigative journalismLow — AI drafts are weakNo documented case of AI replacing long-form investigative capacity
Music composition (background/functional)Medium-HighEpidemic Sound, Soundraw, Udio offer AI music; used in advertising and YouTube extensively by 2024
Film/video VFXGrowing rapidlyRunway Gen-2, Sora; studios using AI for background generation and de-aging in 2024 productions
The Stock Image Collapse

The most complete case of AI disrupting a creative market is stock photography and illustration. The timeline is stark:

2022: Midjourney, DALL-E, Stable Diffusion release. Quality is impressive for many commercial use cases.

2023: Shutterstock announces it will sell AI-generated images and pays a fund to contributors whose work trained the model — implicitly acknowledging their work was used. Getty Images sues Stability AI for scraping its library without compensation.

2024: Multiple independent illustrators report 40–70% income drops from stock platforms. The market for generic commercial illustration — characters, backgrounds, product mockups — has largely shifted to AI generation. The market for distinctive artistic style, editorial illustration, and work requiring a known creator identity has held better but not been unaffected.

Real Case · Sports Illustrated · 2023

In November 2023, Futurism reported that Sports Illustrated had published articles under fake AI-generated author names with AI-generated profile photos. The publisher, The Arena Group, initially claimed the content was from a licensing partner. The revelation caused significant reputational damage and the CEO was subsequently ousted. It illustrated both that AI content was already being published at scale at major outlets — and that the lack of disclosure carried serious consequences.

What Is Holding — and Why

Not all creative work is equally disrupted. The pattern that has emerged:

Most affected: Generic, commodity creative work — stock images, boilerplate ad copy, template-based design, routine data-journalism articles, background music. These were already low-margin, high-volume tasks.

Less affected (so far): Work where the creator's identity, voice, or relationship is the product. A columnist's following is built on their perspective. A brand photographer's clients buy their eye. A novelist's readers want their world. AI cannot replicate established creative identity — yet.

The dangerous middle: Mid-career creatives who specialized in "good enough" commercial work — competent but not distinctive. They are losing ground fastest because AI output is increasingly "good enough" for the use cases they served.

Commodity Creative WorkCreative output produced in volume at low margin for generic commercial uses — stock images, template copy, background music. This category is the most disrupted by generative AI as of 2024.
Identity PremiumThe market value attached to a specific human creator's identity, voice, or relationship with an audience — something AI cannot replicate and which provides some insulation from automation.
The Strategic Lesson

For anyone in a creative field, the question to ask is: Is what I produce valuable because of its content — or because of who made it? Content-value is increasingly replicable by AI. Identity-value is not. Building a distinctive voice, documented expertise, or audience relationship is now a career survival strategy, not just a nice-to-have.

Lesson 3 Quiz

Creative and Media Work — 3 questions
The 2023 WGA writers' strike was historically significant for AI because:
Correct. The WGA deal was a landmark — the first major contract to regulate AI's role in creative work, requiring disclosure and limiting how AI-generated content could be used without proper compensation structures.
The WGA deal did not ban AI outright — it established regulatory language: disclosure requirements, limits on using AI-generated outlines to reduce writer compensation, and protections that were carefully negotiated rather than absolute prohibitions.
According to the lesson, which category of creative workers has been "least affected so far" by AI disruption — and why?
Correct. The "identity premium" — value derived from who made something, not just what it is — provides the strongest insulation. A columnist's audience follows them, not their content format. A photographer's clients want their specific eye.
Stock illustrators are actually among the most disrupted. Mid-career commercial creatives are in the "dangerous middle." Creators with distinctive identity and audience relationships have the strongest insulation — their value is attached to the person, not just the output.
The Sports Illustrated AI author scandal of 2023 revealed which two things simultaneously?
Correct. The case was a two-edged revelation: AI was already mainstream enough that a major publisher was using it at scale — and the backlash showed that lack of transparency had real consequences, including leadership changes.
The scandal showed two things: AI-generated content at scale was already real at a major media brand (not hypothetical), AND that doing it covertly caused serious reputational damage — the CEO was ousted following the revelation.

Lab 3 · Creative Capability Test

Run AI through real creative tasks — then assess what it can and can't do

Your Task

This lab uses the AI as a creative tool — then has you analyze the output critically. You'll test AI's performance across different creative tasks and identify where it produces "good enough" commodity output vs. where human creative identity still wins.

Start with: "Write a 100-word product description for a sustainable water bottle, in an enthusiastic brand voice." Then try: "Now write the same thing but make it sound like David Sedaris wrote it." Compare the outputs. What changes? Where does the AI's identity limitation become visible?
AI Lab Assistant
Creative Analysis Mode
Let's test creative capabilities together. Give me a creative task — copy, story, description, pitch, anything — and I'll produce it. Then we can analyze: Is this "good enough" for commercial use? Does it have an identity? What would a human creator bring that this lacks? Let's find the real limits.
Module 2 · Lesson 4

Physical Work, Retail, and the Jobs That Held

Not all work is text-in, text-out. Physical complexity, social trust, and local presence have proven surprisingly durable.
Which jobs are NOT being disrupted as fast as predicted — and what makes them more resilient than knowledge work?

In 2022 and 2023, major retailers made high-profile bets on automation. Amazon's "Just Walk Out" cashierless technology was deployed in Amazon Fresh stores and licensed to third-party retailers globally. Walmart expanded its use of shelf-scanning robots. Ahold Delhaize deployed robots in Giant Food stores for inventory scanning.

By early 2024, Amazon had quietly removed Just Walk Out technology from all its Amazon Fresh stores, replacing it with "Dash Carts." The reason cited internally: the system relied heavily on human reviewers in India watching video feeds to classify purchases — reportedly more than 1,000 workers. The promised fully-automated checkout had not materialized. Physical retail proved harder to automate than advertised.

The Physical World Is Hard

Robotics and AI face a fundamental challenge in unstructured physical environments. This is sometimes called Moravec's Paradox — named after roboticist Hans Moravec, who observed in the 1980s that the tasks hardest for humans (complex math, chess, language) are easiest for computers, while tasks easy for humans (walking, picking up a cup, recognizing an object in unusual lighting) remain extremely hard for machines.

This paradox has not been fully resolved. LLMs solved the language problem but did not solve the physical-world manipulation problem. The result: jobs requiring dexterous physical work in unpredictable environments have held better than predicted.

Job CategoryAutomation Prediction (circa 2017)Actual Status 2024
Truck driversHighly at risk; autonomous vehicles would replace within a decadeStill in strong demand; autonomous trucking limited to specific routes, not widespread deployment
Fast food workersRobots would replace order-taking and food assemblyKiosks common for ordering; food assembly automation limited and expensive; net employment largely stable
Construction workersModular and robotic construction would reduce labor needsLabor shortages, not surpluses; robotics adoption slow due to site variability
Plumbers, electriciansEventually automatable as robotics advancedSevere shortages; wages rising significantly; no meaningful robotics deployment
Home health aidesUncertain; social interaction seen as barrierFastest-growing US occupation 2023–2024; AI tools assist but do not replace
BLS Data Point

The US Bureau of Labor Statistics Occupational Outlook Handbook (2024 edition) projects home health and personal care aides will grow by 22% from 2022–2032 — adding over 800,000 jobs. This is among the fastest growth of any occupation. Care work requiring sustained human presence, emotional attunement, and physical assistance has proven highly resistant to automation.

The Trades: A Counter-Narrative

In the same period that AI disrupted knowledge work entry-level hiring, skilled trades faced the opposite problem: severe labor shortages. The reasons are structural and long-developing, but the contrast is striking.

Electricians: The US needs approximately 80,000 new electricians per year. Current training pipelines produce far fewer. The electric vehicle transition and data center buildout (driven in part by AI infrastructure) are creating enormous additional demand. The Bureau of Labor Statistics projects electrician employment to grow 11% from 2023–2033 — faster than average for all occupations.

HVAC technicians: Similar story. The transition to heat pumps and more efficient systems requires skilled installation and maintenance. Employment projected to grow 9% through 2033.

Plumbers: An aging workforce is retiring faster than new workers enter. The combination creates wage pressure upward — journeyman plumbers in many US cities earned $80,000–$110,000 in 2024, with master plumbers often exceeding $120,000.

The AI wave is actually increasing demand for some physical trades, because AI infrastructure (data centers) requires massive electrical, HVAC, and plumbing installation.

22%Projected growth in home health aide jobs 2022–2032 (BLS)
11%Projected electrician job growth 2023–2033 (BLS)
80KAnnual electrician shortage in the US (industry estimate, 2024)
What Makes a Job Resilient Right Now

Based on where AI is and where robotics is in 2024, the following job characteristics correlate with lower current disruption risk:

Physical Dexterity in Variable EnvironmentsWork requiring hands-on manipulation in unpredictable settings — construction sites, plumbing, electrical work, surgery. Robotics has not solved this at commercial scale.
Human Presence and TrustCare work, mental health support, personal training, teaching young children — where the human relationship is the service, not incidental to it.
Local, Non-Digitizable ServiceWork that must happen physically in a specific location — installation, repair, delivery of physical goods, emergency response. Cannot be offshored to AI servers.
High-Stakes Judgment Under UncertaintyCrisis situations, complex negotiations, clinical decisions where an error has serious consequences and context is ambiguous. AI assists but liability and accountability remain human.
The Full Picture

The AI disruption story is not "all jobs are at risk." It is more precise: text-based, routine, scalable knowledge tasks at the junior level are the most disrupted right now. Physical, relational, variable-environment, and high-stakes judgment work is holding — and in many cases growing. Career resilience means understanding which side of this line your work sits on, and whether you can move toward the resilient side if needed.

Lesson 4 Quiz

Physical Work, Retail, and the Jobs That Held — 3 questions
Amazon's "Just Walk Out" cashierless technology was removed from all Amazon Fresh stores by early 2024. What did investigations reveal about how the system actually worked?
Correct. This case is a vivid illustration of "AI-washing" — technology marketed as fully automated that turned out to require substantial human labor behind the scenes. Physical retail automation proved far harder than the pitch suggested.
The issue was more fundamental: the "AI" system reportedly required more than 1,000 human reviewers watching video feeds to accurately classify what customers picked up. The promised automation was substantially human-powered, making the economics unworkable.
Moravec's Paradox helps explain why some physical jobs are more automation-resistant than knowledge work. What does it state?
Correct. Moravec's Paradox explains why LLMs solved language before robotics solved "pick up a cup" — the human cognitive hierarchy doesn't match the machine difficulty hierarchy. This is why plumbers are scarce while paralegals face AI competition.
Moravec's Paradox is about the mismatch between human-perceived difficulty and machine-perceived difficulty. What's hard for humans (calculus, language) is often easy for computers. What's easy for humans (walking, recognizing objects in odd lighting, dexterous manipulation) remains very hard for machines.
The lesson notes that AI is actually INCREASING demand for some physical trades. Which mechanism explains this?
Correct. Data centers that run AI systems require enormous physical infrastructure — electrical systems, cooling (HVAC), plumbing, and structural work. The AI wave is creating physical-world demand, not just eliminating physical-world jobs.
The key mechanism is data center buildout. Every AI system runs on physical servers in physical buildings that require electricians, HVAC technicians, and plumbers to build and maintain. AI is simultaneously disrupting knowledge work and creating demand for skilled trades that support its own infrastructure.

Lab 4 · Resilience Assessment

Map your own career position against the disruption patterns in this module

Your Task

This is the most personal lab in Module 2. Use the AI to do a structured resilience assessment of your own career position — or a career you're considering. The goal is to produce an honest, specific assessment: which tasks are exposed, which are resilient, and what moves could increase your resilience.

Start with: "I work as a [job title] and my main tasks include [list 3–5 things you do]. Based on what you know about AI disruption patterns in 2024, help me identify: (1) which of my tasks are most at risk, (2) which are most resilient, and (3) what specific skills or pivots could strengthen my position."
AI Lab Assistant
Career Resilience Mode
Ready for your resilience assessment. Tell me your role and what you actually do day-to-day — be specific about the types of tasks, not just the job title. I'll give you an honest, structured read on exposure and resilience based on what we know about where AI disruption is and isn't happening in 2024.

Module 2 Test

Which Jobs AI Is Changing Now — 15 questions · Pass at 80%
1. IBM CEO Arvind Krishna announced in May 2023 that the company would pause hiring for roughly 7,800 roles. Which types of roles did he specifically cite?
Correct. Krishna specifically named back-office roles — HR, finance, document processing — as the positions expected to be replaced by AI within five years.
Krishna specified back-office roles: HR, finance, and document processing — the most routinizable knowledge work tasks at IBM.
2. Goldman Sachs' internal analysis estimated that generative AI could automate what proportion of work tasks at legal, accounting, and consulting firms?
Correct. 25–50% — a significant but not total displacement estimate, focused on specific task categories like document review, first-draft writing, and data summarization.
Goldman's estimate was 25–50% of work tasks — a substantial fraction but not total replacement, concentrated in routine knowledge tasks.
3. BT Group announced it would cut 55,000 jobs by 2030 and specifically cited AI. Which functions were most associated with this reduction?
Correct. BT's AI-driven reductions targeted back-office, monitoring, and administrative functions — the same pattern seen across other enterprises.
BT cited back-office functions — network monitoring, billing queries, HR ticketing — as the areas where AI would enable headcount reduction.
4. In the task-level framework, "job polarization" refers to:
Correct. Job polarization describes the hollowing of the middle — which has been happening since 2000 and is now being extended upward into professional knowledge work by AI.
Job polarization is the labor market phenomenon where mid-skill, middle-income roles shrink while extremes (very high-skill and low-skill physical) hold — AI is extending this pattern into professional knowledge work.
5. Allen & Overy's firm-wide deployment of Harvey AI in 2023 primarily impacted which types of legal work?
Correct. Harvey handled first-draft contracts, client questions answered from prior case data, and preliminary regulatory research — classic junior associate tasks.
Harvey AI at Allen & Overy handled first-draft contracts, client questions from case data, and preliminary research — the routine text-generation tasks traditionally done by junior associates.
6. GPT-4 scored in what percentile on the bar exam when tested in 2023?
Correct. GPT-4 scored in the 90th percentile — well above passing and well above average human test-takers. This illustrates that AI has achieved credentialed-knowledge-level performance, while judgment-under-ambiguity remains more resistant.
GPT-4 scored in the 90th percentile on the bar exam — demonstrating that credentialed knowledge retrieval is no longer sufficient differentiator from AI, only judgment and relationships are.
7. DeepMind's breast cancer detection AI was notable because its results showed:
Correct. The DeepMind results showed AI outperforming the average radiologist on mammogram screening — a significant milestone showing AI achieving above-human performance on a specialized medical task.
DeepMind's AI detected breast cancer with greater accuracy than the average radiologist — fewer false negatives AND fewer false positives — a meaningful clinical performance benchmark.
8. The 2023–2024 Upwork hiring data study (Harvard/University of Chicago) found which category had the SECOND largest decline after data entry?
Correct. Writing and content creation fell 21% — second only to data entry and research (−29%), and ahead of basic coding (−17%).
Writing and content creation saw a −21% decline — the second largest drop after data entry/research (−29%), followed by basic coding (−17%).
9. The 2023 WGA writers' strike resulted in contract language that:
Correct. The WGA deal was pragmatic, not prohibitive — disclosure requirements and limits on using AI output to reduce writer pay, not an outright ban.
The WGA agreement required disclosure and set limits on AI use reducing writer compensation — a regulatory framework, not a ban, and the first of its kind in a major US labor contract.
10. The "identity premium" in creative work refers to:
Correct. Identity premium is the insulation effect when your value is tied to who you are, not just what you produce. Audiences follow columnists, clients hire specific photographers, readers buy authors — AI can't replicate that relational dimension.
The identity premium is the market value of being a specific, known creator — where value is attached to the person, not just the output. AI can replicate output; it cannot replicate established identity and trust relationships.
11. Amazon's "Just Walk Out" technology was removed from Amazon Fresh stores. What was the key reason cited?
Correct. This case illustrates "AI-washing" — marketing automation that is actually hybrid human-AI labor. The human review requirement made the economics unworkable.
The system required over 1,000 human reviewers in India to watch video footage and classify purchases — meaning the "fully automated" checkout was substantially human-powered, making it both misleading and economically unviable.
12. Moravec's Paradox predicts that which category of jobs would be MORE resistant to near-term automation?
Correct. Physical manipulation in variable environments (what humans find trivially easy) remains extremely hard for machines — the core of Moravec's Paradox and why plumbers are scarcer than paralegals facing AI competition.
Moravec's Paradox says physical manipulation in unpredictable environments — what humans find easy — is hardest for machines. Mathematical reasoning, language, and pattern matching are the areas where computers excel.
13. The BLS projects home health aide jobs to grow by what percentage from 2022–2032?
Correct. 22% growth, adding 800,000+ jobs — one of the fastest growth rates of any occupation, driven by aging demographics and the fundamental resistance of care work to automation.
BLS projects 22% growth for home health aides — one of the fastest rates of any occupation, adding over 800,000 jobs. Care work requiring sustained human presence is highly automation-resistant.
14. The lesson explains that AI is paradoxically INCREASING demand for electricians, HVAC technicians, and plumbers. What is the mechanism?
Correct. AI runs on physical hardware in physical buildings. Data center buildout — driven by AI compute demand — is creating significant demand for the trades needed to build and maintain that infrastructure.
The mechanism is data center infrastructure: AI compute runs on physical servers requiring massive electrical systems, HVAC cooling, and plumbing. The AI wave creates demand for the physical trades that build its own infrastructure.
15. Based on all four lessons, which statement BEST summarizes where AI disruption is most concentrated in 2024?
Correct. This is the precise, evidence-based summary: AI disruption in 2024 is concentrated in routine knowledge tasks at junior levels, while physical, relational, and judgment-intensive work remains resilient or is growing.
The clearest pattern from real documented cases: routine knowledge tasks (document review, first drafts, data research, customer service text) at junior levels are most affected; physical, relational, and high-stakes judgment work is holding or growing. AI disruption is concentrated, not uniform.