The Structural Squeeze: Artificial Intelligence, Entry-Level Labor Dynamics, and the Escalation of Generational Economic Exclusion

Full report

On this page

1. Introduction and Analytical Framework: The Re-engineering of the Entry-Level Knowledge Economy

The integration of generative artificial intelligence and agentic software systems into the white-collar economy marks a structural shift in how young adults enter the workforce, gain professional competence, and build wealth1. Historically, technological innovations primarily automated physical labor or repetitive, non-cognitive administrative tasks3. In contrast, modern artificial intelligence models target cognitive processing, information synthesis, introductory computer programming, routine legal drafting, and basic quantitative analysis2. These cognitive tasks have served for decades as the foundational training ground for early-career professionals across knowledge-intensive industries2. By automating the routine tasks traditionally assigned to entry-level staff, employers are refactoring corporate workforce structures4. This trend risks pulling up the ladder of career advancement8. Organizations can maintain or expand output using smaller, senior-heavy teams paired with autonomous AI tools, reducing their reliance on entry-level hiring9. While aggregate macroeconomic indicators show steady overall employment, a structural divergence is emerging beneath the surface2. The traditional entry-level career pathway—where recent graduates trade routine labor for mentorship, skill development, and upward income mobility—is experiencing significant compression6. To analyze this shift, labor economics distinguishes between task-level automation and role-level displacement7. The complete elimination of an entire occupation is rare; instead, roles undergo workflow re-engineering4. However, when routine tasks make up 50% to 60% of an entry-level position's workload, automating those tasks sharply reduces the need for entry-level headcount4. This investigation examines the mechanics of this contraction across major professional sectors, separates technological impacts from broader economic trends, assesses the consequences for human capital development, and outlines policy responses to prevent long-term economic exclusion among younger generations11.

2. Early-Career Employment Trends and Graduate Recruitment (2018–2026)

The early-career labor market in the United States experienced significant volatility between 2018 and late 20262. Before 2020, entry-level recruitment followed predictable cyclical patterns anchored by campus career fairs, structured corporate training pipelines, and steady conversion rates from internships to full-time roles17. The COVID-19 pandemic caused a brief disruption in early 2020, followed by an aggressive expansion in tech, finance, and professional services through 2021 and 20222. Driven by zero-interest-rate policies and accelerated digital adoption, employers hired heavily, pushing entry-level compensation and hiring volumes to record highs2. This expansion reversed sharply between 2023 and 20252. As monetary policy tightened and businesses shifted their focus toward operational efficiency, white-collar entry-level hiring slowed significantly2. Data from the National Association of Colleges and Employers (NACE) tracks this shift14. Initial employer projections for the college Class of 2026 gathered in late 2025 indicated flat hiring intentions, with a plurality of recruiters rating the job market as merely "fair"—a sentiment not seen since the economic disruption of 202114. By the spring of 2026, employer intentions split across sector lines18. While total entry-level hiring projections adjusted upward to a 5.6% expansion—driven primarily by non-tech sectors such as engineering services, construction, wholesale trade, and healthcare—professional knowledge roles in software, corporate finance, and media remained constrained11.

Simultaneously, employer selection criteria underwent structural changes19. By 2026, 70.5% of employers reported adopting skills-based hiring frameworks over strict GPA or institutional filtering19. Employer priorities shifted toward verified technical proficiencies, problem-solving capabilities (rated essential by 90% of employers), teamwork (80%), and applied internship experience24. The traditional assumption that a university degree alone guarantees entry into professional career ladders has been dismantled11.

3. Sectoral Breakdown of Entry-Level Vulnerabilities and Empirical Evidence

The impact of artificial intelligence deployment on entry-level employment varies across professional domains4. Occupations heavy in structured data processing, routine code writing, document review, and standardized communication display high vulnerability to early-career task replacement4.

Junior Software Engineers and Developers

The software engineering sector offers a clear case study of entry-level workforce restructuring10. Following the widespread adoption of AI coding assistants and autonomous development frameworks, employer demand for software engineers inverted10. Analysis of job posting data from Lightcast and IZA indicates a 14% to 15% relative decline in junior software developer vacancies compared to senior roles following the deployment of LLM coding tools28. Data from the Stanford Canaries Dashboard reveals that the gap between entry-level and senior developer posting growth expanded from 15% in mid-2025 to 19% by mid-202630. SignalFire’s 2026 labor study confirmed that new-graduate engineering hiring fell by approximately 65% at major technology firms and 76% at early-stage startups compared to 2019 baselines10. While overall developer employment did not collapse, hiring composition shifted significantly10. Between May 2025 and May 2026, senior developer titles accounted for 71% of all new software posting growth, with 37% explicitly requiring AI architecture skills16. Total compensation gradients expanded accordingly: Levels.fyi data shows staff engineer median compensation increasing by 7.52% to \$457,000, while entry-level engineer compensation grew by only 1.64% to \$155,00010. In corporate IT prioritization, application development fell off the CIO "hardest-to-fill" list entirely by mid-2026, replaced at the top by AI/Machine Learning integration and cybersecurity10.

Financial Services and Corporate Banking

In corporate finance, investment banking, and financial analysis, generative tools automated pitch deck generation, financial model population, and preliminary earnings transcript synthesis4. These tasks traditionally occupied 60% to 80% of a first-year analyst's workweek4. Financial institutions have begun restructuring intake cohorts, shifting junior roles from manual financial modeling toward output evaluation—testing assumptions, probing AI outputs for logical flaws, and performing qualitative risk analysis4.

Legal Assistants and Paralegals

In legal services, domain-specific AI platforms altered document discovery, due diligence, and legal research32. Tasks that previously required teams of junior associates and paralegals—such as sorting contract clauses or summarizing deposition transcripts—are now executed in minutes32. Consequently, law firms have curtailed entry-level paralegal positions, raising the hiring bar for junior associates to demonstrate immediate client-facing and strategic value7.

Customer Support and Administrative Operations

Customer support and office administration represent direct instances of operational task replacement5. Conversational AI agents capable of resolving multi-step support tickets reduced the need for human tier-1 support staff33. Census Bureau Business Trends and Outlook Survey (BTOS) data illustrates that while firm-level AI adoption tripled to \~15% by early 2026, administrative support occupations experienced steady headcount rationalization11.

4. Deconstructing Confounders: Isolating Technological Effects from Macroeconomic and Policy Drivers

Attributing the entire contraction in entry-level white-collar hiring solely to artificial intelligence oversimplifies complex labor market mechanics31. To isolate the true displacement effect of AI, four major macroeconomic and regulatory confounders must be evaluated12.

Federal Reserve Monetary Tightening and Interest Rate Cycles

Between March 2022 and late 2023, the Federal Reserve raised the federal funds rate from near zero to over 5.25%. This policy shift elevated the cost of capital, disproportionately impacting technology startups and venture-backed enterprise software firms that rely on future cash flow valuations31. As capital expenditures were curtailed, tech firms prioritized immediate cash flow over long-term talent cultivation, slowing entry-level hiring across the sector10.

Post-Pandemic Hyper-Hiring Corrections

During 2021 and early 2022, technology and professional services firms over-hired under the assumption that pandemic-era digital consumption patterns would persist indefinitely2. The subsequent corporate restructuring in 2023 and 2024 served as an inventory correction in human capital2. Research indicates that more than half of the drop in software job postings from the 2022 peak occurred prior to the widespread commercial deployment of generative AI tools in enterprise workflows31.

Federal Tax Policy Dynamics: Tax Code Section 174 Capitalization

A critical driver of the software engineering hiring slowdown was a change in federal tax legislation12. Under the Tax Cuts and Jobs Act (TCJA) of 2017, changes to Internal Revenue Code (IRC) Section 174 took effect for tax years beginning after December 31, 202112. Previously, businesses could immediately expense 100% of domestic software development and R\&D payroll costs12. The TCJA amendment mandated that domestic R\&D expenses be capitalized and amortized over five years (using a mid-year convention), while foreign R\&D had to be amortized over 15 years12. For early-stage technology companies where engineering salaries constitute the primary operating expense, Section 174 created severe taxable income distortions12. A Series A enterprise with \$3 million in engineering payroll could no longer deduct \$3 million in year one, but only \$300,000, creating unexpected six-figure tax liabilities for pre-profit firms12. This tax change led directly to software engineering hiring freezes and layoffs between 2022 and 202412. Legislative resolution occurred on July 4, 2025, with the passage of the One Big Beautiful Bill Act (OBBBA)35. OBBBA created IRC Section 174A, which restored immediate expensing for domestic R\&E expenditures incurred after December 31, 2024, and offered transition rules allowing businesses to accelerate unamortized 2022–2024 domestic balances35. However, Section 174A maintained the mandatory 15-year amortization rule for foreign R\&D expenditures35. The passage of Section 174A removed a major tax headwind for domestic software engineering hiring in 2025 and 202636. The remaining weakness in junior engineering positions can thus be attributed more directly to structural workflow changes driven by AI rather than tax distortion10.

Offshoring and Globalized Labor Arbitrage

The shift toward remote work facilitated international hiring in lower-cost regions12. However, Section 174A’s preservation of the 15-year amortization penalty for foreign R\&D offset some of this pressure by making foreign engineering teams three times more tax-expensive over time than domestic staff12.

Demographic Shifts and University Enrollment Trends

Demographic trends show higher education enrollment growth leveling off, alongside shifts toward STEM majors39. The supply of computer science graduates expanded rapidly between 2018 and 2024, reaching nearly 94,000 bachelor's degrees annually in the U.S.39. This surge in supply intersected directly with AI-driven early-career hiring slowdowns, compounding graduate underemployment10.

5. The Erosion of Human Capital Pipelines: Apprenticeship, Mentorship, and Learning-by-Doing

The most significant long-term risk posed by entry-level task automation is the disruption of human capital development6. Economic literature on skill acquisition emphasizes the concept of "learning-by-doing"8. Senior expertise is rarely acquired solely through formal education; it is built through years of performing routine tasks, observing senior colleagues, and resolving low-stakes problems6.

Tacit Knowledge Loss

Routine tasks allow entry-level workers to internalize domain patterns, code structures, legal precedents, and financial nuances6. Bypassing these tasks deprives junior staff of foundational tacit knowledge8. When routine execution is handled by automated systems, early-career employees miss out on the repetitive practice that builds deep domain intuition6.

The Novice Reliance Trap

A study published in Science demonstrated that while generative AI boosts text production output by up to 40%, novice workers who accept machine outputs uncritically perform worse on complex problem-solving than unassisted peers4. Without a solid foundation built through basic execution, junior workers struggle to evaluate machine output quality or detect subtle errors4.

The Senior Talent Shortage Paradox

Failing to train junior employees today creates a shortage of senior architects, partners, and managers tomorrow10. Employers face a structural dilemma: they demand experienced senior talent while reducing the entry-level positions required to produce that talent10. By automating the lower rungs of the professional career ladder, organizations capture short-term productivity gains at the expense of long-term workforce sustainability6. In response, progressive organizations are redesigning entry-level workflows4. In accounting and investment management, firms are replacing manual compilation with output interrogation, prompt engineering, and scenario evaluation4. In software engineering, code review, system architecture, and security validation are moving earlier into junior career paths10.

6. Macroeconomic Cascades: Underemployment, Credential Inflation, and Household Delay

The narrowing of entry-level white-collar pathways has escalated underemployment and structural financial pressures among young college graduates11. Data from the Federal Reserve Bank of New York indicates that while the unemployment rate for recent college graduates (ages 22–27) remained steady at 5.6% in 2026, the underemployment rate reached 42.0%20. Furthermore, the quality of underemployed jobs has deteriorated25. Fifteen years ago, 47% of underemployed recent graduates secured relatively high-paying non-college positions earning above \$45,000 annually; by 2026, that share declined to 34%25. Labor market outcomes diverge significantly by academic discipline, as detailed in the New York Fed's graduate tracking data39:

Socioeconomic Downstream Cascades

This underemployment pressure creates compounding socioeconomic disruptions across young adult populations39:

• Credential Inflation: As traditional bachelor-level roles automate, employers demand master's degrees or specialized microcredentials for entry-level roles26. This raises time and financial requirements without guaranteeing proportional wage premiums41.
• Debt Servicing and Wealth Accumulation Constraints: High student debt obligations paired with stagnant entry-level wages reduce disposable income, delaying homeownership and personal investment41.
• Delayed Household Formation: Early-career income instability correlates directly with delayed home acquisitions, later marriage ages, and declining fertility rates among educated young adults41.

7. Socioeconomic Segmentation and Demographic Inequities

The macroeconomic impact of AI adoption is non-uniform, exposing key socioeconomic and demographic fault lines across the labor force11.

The Non-Degree vs. College-Educated Labor Market Inversion

A notable labor market trend between 2024 and 2026 is the decoupling of outcomes between college-educated and non-college-educated workers11. Data synthesized by the Burning Glass Institute shows that 88 million U.S. workers aged 22 and over without a bachelor's degree experienced near 20-year low unemployment rates11. Conversely, young college graduates aged 22 to 34 saw their unemployment metrics weaken, placing them at the 70th percentile of their historical unemployment range since 200311. This trend represents a historical inversion11. Physical labor, healthcare support, skilled trades, construction, and specialized services maintain structural demand headroom and high scalability barriers that protect them from automation7. Meanwhile, entry-level cognitive roles face direct technological substitution7.

Socioeconomic Buffer Capacity

The contraction of entry-level professional roles creates class-based divergence among graduates8:

• High-Resource Graduates: Young adults from affluent families rely on parental financial support, enabling them to accept unpaid internships, pursue additional certifications, or wait out extended job searches8.
• Low-Resource and First-Generation Graduates: Graduates lacking family financial buffers must accept immediate non-college underemployment (e.g., in retail, hospitality, or logistics) to meet debt and living expenses21. This increases the risk of long-term career under-matching40.

Geographic Concentration

White-collar entry-level hiring has concentrated further in major economic centers (e.g., Austin, Seattle, New York, San Francisco)26. While offering higher hiring volumes, these cities carry elevated living costs26. The reduction of fully remote entry-level roles isolates rural and non-metropolitan graduates from early-career opportunities24.

8. Critical Evaluation of AI-Native Occupations: Career Pathways vs. Experience Barriers

A central debate in labor economics is whether new AI-centered occupations can absorb displaced entry-level white-collar workers2. Proponents argue that the expansion of artificial intelligence creates new occupational categories, such as prompt engineers, AI quality assurance evaluators, model fine-tuners, and enterprise AI integration consultants19. Empirical analysis indicates that these AI-native roles do not currently function as direct substitutes for traditional entry-level pathways19:

High-Skill AI Occupations

Roles such as Machine Learning Engineer, Model Architect, and AI Infrastructure Developer command high pay (\$140,000 to \$190,000+ starting compensation)19. However, these roles require advanced mathematical skills, software engineering fundamentals, and deep technical competencies19. They do not offer an accessible entry point for general business, humanities, or junior administrative candidates19.

Enterprise Integration and Compliance Roles

Positions focusing on AI workflow integration, algorithmic auditing, and regulatory compliance demand deep domain knowledge43. Employers seeking AI integration specialists generally hire experienced professionals who understand existing corporate operations rather than recent university graduates43.

Low-Skill AI Support Occupations

Entry-level AI roles such as data labelers, content annotators, and output evaluators offer low entry barriers43. However, these positions frequently feature gig-economy structures, low wages, limited job security, and minimal opportunities for professional advancement43. They do not provide the structured mentorship or skill accumulation characteristic of traditional professional entry-level roles6. Consequently, while AI creates technical positions, these roles primarily benefit workers who already possess technical experience or specialize in advanced quantitative fields3.

9. Political Economy: Alienation, Institutional Distrust, and Historical Parallels

The structural blockage of early-career economic advancement carries political and civic risks43. Political economy literature documents that periods of educational expansion paired with shrinking professional opportunities generate social and political friction41. Comparing contemporary labor dynamics with historical periods of youth economic exclusion provides valuable context:

• The 1930s Great Depression: Mass youth unemployment disrupted economic life, prompting major structural labor reforms, the establishment of the Civilian Conservation Corps, and federal public employment interventions.
• The 1970s Stagflation Era: High inflation paired with slowing growth left university graduates facing credential misalignment, fostering widespread institutional cynicism.
• The Post-2008 Global Financial Crisis: In the aftermath of the GFC, young adults in Western economies faced prolonged unemployment and underemployment. This economic exclusion contributed to political movements across the political spectrum, fueled by distrust in financial and political institutions.
• The 2020s AI Knowledge Squeeze: Unlike previous crises that primarily impacted non-college or blue-collar workers, the modern squeeze directly affects the educated young population11. These individuals invested significant capital into formal education under the premise of meritocratic advancement41. Facing underemployment, student debt, and unaffordable housing, this cohort displays growing disillusionment with traditional political and economic systems41.

This dynamic fuels political polarization, systemic distrust in institutional expertise, and support for economic nationalism or disruptive political movements43. When young adults are excluded from economic participation, their commitment to political stability weakens43.

10. Long-Term Macroeconomic Scenarios Through 2040

To assess the long-term impact of artificial intelligence on entry-level employment, three distinct macroeconomic scenarios are modeled through 2040:

Scenario A: Task Augmentation and Re-Skilling Rebound

Under this optimistic scenario, productivity gains from AI lower production costs across knowledge-intensive sectors7. Elastic demand for software, legal services, and financial analysis expands total market output7. Employers redesign entry-level roles around AI management, systems architecture, output validation, and client relationship building4. While pure task-execution positions decline, overall early-career employment recovers as new business models emerge43. The productivity dividend generates new employment opportunities, maintaining intergenerational social mobility43.

Scenario B: The Closed Ladder and Structural Seniority Trap

Under this stagnation scenario, AI adoption follows a "so-so automation" model, where algorithms replace human labor without producing massive aggregate productivity gains or new demand elasticity9. Enterprise hiring freezes at the entry level become permanent10. Senior workers use AI to manage operations, but the human capital pipeline collapses6. By 2035, firms face severe shortages of experienced senior talent10. Underemployment stabilizes at high levels (\>50%), wealth inequality widens between older asset holders and younger labor sellers, and generational economic mobility stalls9.

Scenario C: Neo-Guild Institutional Re-engineering

Under this interventionist scenario, public policy, tax reforms, and labor regulations mandate structural changes in corporate workforce development13. The state penalizes complete early-career workforce elimination through target tax structures while incentivizing formal white-collar apprenticeships13. Higher education shifts from four-year knowledge-retention models to co-op and industry-integrated training structures26. Institutional intervention re-establishes entry-level career pathways, distributing the productivity dividend through policy45.

11. Sourced Policy Proposals and Institutional Reforms

Mitigating generational economic exclusion requires structural tax, labor, and educational policy interventions13:

1. Targeted Tax Incentives and Early-Career Hiring Credits

Legislative frameworks have emerged to address white-collar labor displacement13:

• The SKILL Act (June 2026): Introduced by Representative Liccardo, the SKILL Act establishes tax credit incentives for employers participating in workforce development13. The bill provides employers with a \$2,500 tax credit for each student completing a qualifying training program, and an additional \$2,500 credit for each graduate hired into a full-time role, supported by \$500 million in annual state-allocated credit authority13.
• The Gottheimer-Lawler AI Workforce Proposal: Introduced by Representatives Josh Gottheimer and Mike Lawler, this legislation proposes a 30% tax credit for businesses investing in AI workforce training, upskilling, and early-career retention initiatives46.
• Tax Code Optimization: Following the passage of OBBBA Section 174A in July 2025 (which restored domestic R\&D expensing), tax policy can be further aligned by establishing preferential deduction rates for firms that maintain defined entry-level and apprentice intake ratios35.

2. Workforce Innovation and White-Collar Apprenticeship Frameworks

• WIOA Expansion: Updating the Workforce Innovation and Opportunity Act (WIOA) to raise the cap on incumbent worker training and extend Registered Apprenticeship frameworks from industrial trades into white-collar professions (e.g., cybersecurity, software development, paralegal services, financial analysis)45.
• Public-Private Apprenticeship Subsidies: Offering federal wage-matching subsidies for small and mid-sized enterprises that hire and train junior professionals in AI-impacted sectors45.

3. Higher Education Curriculum Transformation

University curricula must adapt to the presence of automated execution tools4:

• Shift to Critical Framing and Red-Teaming: Universities must transition away from grading students on routine synthesis, basic coding exercises, or standardized report writing—tasks now performed instantly by AI models4. Higher education must focus on teaching critical framing, system architecture, prompt engineering, and output interrogation4.
• Mandatory Co-op and Applied Experiential Learning: Higher education institutions must embed multi-semester cooperative work placements directly into undergraduate degree programs26. As degree credentials alone lose signaling power, applied experience becomes essential for graduate employment26.

12. Conclusion: Task Automation vs. The Intergenerational Advancement Mechanism

The central question facing labor economists and policymakers is whether artificial intelligence threatens only specific entry-level tasks or destabilizes the broader intergenerational process of socio-economic mobility4. The evidence indicates that the threat extends beyond task-level automation4. While claims of an immediate, aggregate job collapse are unsupported by macro data, focusing solely on top-line unemployment figures overlooks a critical structural shift2. Generative artificial intelligence is altering the entry-level white-collar labor ecosystem4. By automating the routine execution tasks that historically formed the foundation of early-career employment, AI removes essential components of the human capital development pipeline6. The immediate efficiency gains realized by corporate enterprises carry clear systemic trade-offs:

1. The operational pathways that transform recent university graduates into skilled senior professionals are narrowing6.

2. College graduates face elevated underemployment rates (42.0%), qualitative wage deterioration, and delayed wealth accumulation20.

3. The economic return on higher education is bifurcating, rewarding specialized technical majors while eroding options for broader knowledge disciplines11.

4. Prolonged economic exclusion among educated young adults risks fostering political alienation, institutional distrust, and civic instability43.

Artificial intelligence does not render human labor obsolete7. However, left unmanaged by public policy and corporate strategy, its unguided deployment threatens to sever the traditional operational ladder of economic advancement8. Preserving intergenerational mobility requires restructuring workforce development13. Employers, educational institutions, and policymakers must actively shift from treating entry-level staff as costs to be automated, toward designing structured entry-level roles centered on critical evaluation, system design, and human judgment4.

Works cited

1. Artificial Intelligence and the Future of Entry-Level Work, https\://reports.weforum.org/docs/WEF\_Artificial\_Intelligence\_and\_the\_Future\_of\_Entry\_Level\_Work\_2026.pdf

2. What is really happening to jobs? Separating AI hype from reality, https\://siepr.stanford.edu/publications/policy-brief/what-really-happening-jobs-separating-ai-hype-reality

3. Firm Investments in Artificial Intelligence Technologies and Changes, https\://www\.nber.org/system/files/working\_papers/w31325/w31325.pdf

4. The Perils of Using AI to Replace Entry-Level Jobs, https\://hbsp.harvard.edu/inspiring-minds/ai-impact-entry-level-jobs

5. Early Labor Market Transformation under Generative AI \- NBER, https\://www\.nber.org/system/files/working\_papers/w33777/w33777.pdf

6. Does AI Assistance Enhance or Erode Expertise? Evidence ... \- NBER, https\://www\.nber.org/system/files/working\_papers/w35720/w35720.pdf

7. AI Will Reshape More Jobs Than It Replaces | BCG, https\://www\.bcg.com/publications/2026/ai-will-reshape-more-jobs-than-it-replaces

8. GETTING IN THE FRONT DOOR IN THE AGE OF AI: \- IE University, https\://static.ie.edu/CGC/CGC\_FrontDoorAI\_Lipson.pdf

9. AI and Jobs: Evidence from Online Vacancies \- NBER, https\://www\.nber.org/system/files/working\_papers/w28257/revisions/w28257.rev1.pdf

10. The Software Engineer Shortage in 2026: What AI Changed, https\://hatchworks.com/blog/gen-ai/software-engineer-shortage/

11. All Sources | jobsdata.ai Research Library, https\://jobsdata.ai/research

12. Section 174 R\&D Amortization: What Startups Need to Know (2026), https\://acuity.co/section-174-changes/

13. Rep. Liccardo Introduces SKILL Act to Prepare Workers for AI-Driven, https\://liccardo.house.gov/media/press-releases/rep-liccardo-introduces-skill-act-prepare-workers-ai-driven-job-market-shifts

14. Hiring Flat for the College Class of 2026 \- NACE, https\://www\.naceweb.org/about-us/press/hiring-is-flat-for-the-college-class-of-2026

15. Hiring Flat for the College Class of 2026 \- NACE, https\://www\.naceweb.org/job-market/trends-and-predictions/hiring-flat-for-the-college-class-of-2026

16. AI coding tools are spreading, but the software-job rebound is, https\://mlq.ai/news/ai-coding-tools-are-spreading-but-the-software-job-rebound-is-concentrated-among-senior-roles/

17. Young Worker Ladder Shift | Case Studies | Economy for Everyone, https\://economyforeveryone.org/case-studies/young-worker-ladder-shift/

18. Outlook Brightens for College Class of 2026 Entry-Level Hiring \- NACE, https\://www\.naceweb.org/about-us/press/2026/outlook-brightens-for-college-class-of-2026-entry-level-hiring

19. 7 Highest-Paying Entry-Level Jobs That Are Also Hiring (+ Verified, https\://checksalaries.com/resource-hub/highest-paying-entry-level-jobs

20. College grad unemployment steady at 5.6% in March, NY Fed reports, https\://www\.investing.com/news/economy-news/college-grad-unemployment-steady-at-56-in-march-ny-fed-reports-93CH-4659630

21. Gen Z Statistics 2026: AI, Jobs, Money and Social Media, https\://www\.usecarly.com/blog/gen-z-statistics/

22. The Damage When Smart People Miss Critical Insights? A \$1.7, https\://shivamber.com/the-damage-when-smart-people-miss-critical-insights-a-1-7-trillion-blind-spot/

23. Software Engineer Career Statistics for 2026: Job Market, Salary, https\://onehour.digital/blog/software-engineer-career-statistics

24. published november 2025 \- 2026 \- NACE, [https\://www\.naceweb.org/docs/default-source/default-document-library/2025/publication/research-report/nace-2026-job-outlook-report.pdf?ct=YTo2OntzOjY6InNvdXJjZSI7YToyOntpOjA7czoxNDoiY2FtcGFpZ24uZXZlbnQiO2k6MTtpOjI1OTt9czo1OiJlbWFpbCI7aToyNTk5O3M6NDoic3RhdCI7czoyMjoiNmE4NmE0MWM4NDI2YzY1OTM0NTU1OSI7czo5OiJzZW50X3RpbWUiO2k6MTc4NzIwODczMjtzOjQ6ImxlYWQiO3M6NzoiODQ2MTY3MSI7czo3OiJjaGFubmVsIjthOjE6e3M6NToiZW1haWwiO2k6MjU5OTt9fQ%3D%3D\&utm\_source=campaign-email-1-slide-deck\&utm\_medium=email\&utm\_campaign=job-outlook-2026](https://www.naceweb.org/docs/default-source/default-document-library/2025/publication/research-report/nace-2026-job-outlook-report.pdf?ct=YTo2OntzOjY6InNvdXJjZSI7YToyOntpOjA7czoxNDoiY2FtcGFpZ24uZXZlbnQiO2k6MTtpOjI1OTt9czo1OiJlbWFpbCI7aToyNTk5O3M6NDoic3RhdCI7czoyMjoiNmE4NmE0MWM4NDI2YzY1OTM0NTU1OSI7czo5OiJzZW50X3RpbWUiO2k6MTc4NzIwODczMjtzOjQ6ImxlYWQiO3M6NzoiODQ2MTY3MSI7czo3OiJjaGFubmVsIjthOjE6e3M6NToiZW1haWwiO2k6MjU5OTt9fQ%3D%3D&utm_source=campaign-email-1-slide-deck&utm_medium=email&utm_campaign=job-outlook-2026)

25. New York Fed Highlights Underemployment Among College, https\://www\.aei.org/commentary/new-york-fed-highlights-underemployment-among-college-graduates/

26. 2027 Entry-Level Hiring Trends for Engineering Graduates, https\://research.com/degrees/entry-level-hiring-trends-for-engineering-graduates-where-new-grads-have-the-best-odds

27. What Are Transferable Skills? Find Yours (2026 Guide) | Extern, https\://www\.extern.com/post/transferable-skills-student-guide

28. Generative AI and the Redefinition of Entry-Level Software Work, https\://www\.iza.org/publications/dp/18723/generative-ai-and-the-redefinition-of-entry-level-software-work

29. Generative AI and the Redefinition of Entry-Level Software Work, https\://ideas.repec.org/p/iza/izadps/dp18723.html

30. The loudest warning about AI and jobs yet \- Platformer News, https\://www\.platformer.news/ai-jobs-warning-brynjolfsson-acemoglu/

31. Junior Developer Jobs and AI in 2026: What the Hiring Data Really, https\://medium.com/@syamkumar0418/junior-developer-jobs-and-ai-in-2026-what-the-hiring-data-really-says-and-how-to-get-hired-021389af4f40

32. Early labor market transformation under generative AI | Brookings, https\://www\.brookings.edu/articles/still-waters-rapid-currents-early-labor-market-transformation-under-generative-ai/

33. Top 20+ Predictions from Experts on AI Job Loss \- AIMultiple, https\://aimultiple.com/ai-job-loss

34. Redesigning Jobs with LLMs to Enhance Productivity \- arXiv, https\://arxiv.org/html/2512.05659v1

35. Section 174 R\&D Capitalization Explained 2026 \- Indinero, https\://www\.indinero.com/blog/section-174-rd-capitalization-explained-2026/

36. What the end of Section 174 means for software developer hiring, https\://leaddev.com/hiring/end-section-174-means-software-developer-hiring

37. Navigating Section 174 research & experimental expenditures after, https\://kaufmanrossin.com/blog/navigating-section-174-research-experimental-expenditures-after-the-one-big-beautiful-bill-act/

38. What will be the impact of Section 174 in 2026? \- Thomson Reuters, https\://www\.thomsonreuters.com/en/institute/articles/section-174-future

39. Recent Graduate Outcomes by Field of Study \- Opportunity Data, https\://opportunitydata.org/grad-labor-explorer.html

40. The Jobs and Degrees Underemployed College Graduates Have, https\://www\.stlouisfed.org/open-vault/2025/aug/jobs-degrees-underemployed-college-graduates-have

41. 2027 Entry-Level Hiring Trends for Computer Engineering Graduates, https\://research.com/degrees/entry-level-hiring-trends-for-computer-engineering-graduates-where-new-grads-have-the-best-odds

42. AI, Productivity, and Labor Markets: A Review of the Empirical, https\://laweconcenter.org/resources/ai-productivity-and-labor-markets-a-review-of-the-empirical-evidence/

43. The AI Labor Debate: Three Views on the Future of Work, https\://carnegieendowment.org/research/2026/04/the-ai-labor-debate-three-views-on-the-future-of-work

44. Is AI Taking Our Jobs? Signal Boosting Noah Smith, https\://economistwritingeveryday.com/2026/10/10/is-ai-taking-our-jobs-signal-boosting-noah-smith/

45. Future of Work in the Age of Automation, Augmentation, and Agentic AI, https\://www\.hks.harvard.edu/sites/default/files/2026-07/Final\_AWP\_276%20FINAL.pdf

46. AI and the Workforce \- Leadership Now, https\://leadershipnowproject.org/wp-content/uploads/2026/05/AI\_Workforce\_v2\_Formatted.docx-2.pdf

47. Cyber Apprenticeship Program | Employer's Guide \- QuickStart, https\://www\.quickstart.com/blog/cyber-workforce-development/building-cyber-apprenticeship-program/

48. The Human Advantage Playbook \- Chamber of Progress, https\://progresschamber.org/project/democratic-ai-opportunity-agenda/human-advantage-playbook/

49. A Future That Works: JFF's Policy Priorities for an AI-Ready Workforce, https\://www\.jff.org/blog/a-future-that-works-jffs-policy-priorities-for-an-ai-ready-workforce/