Agentic AI and Occupational Displacement: A Multi-Regional Task Exposure Analysis of Emerging Labor Market Disruption

Ravish Gupta; Saket Kumar

eess.SYMar 31, 2026

Agentic AI and Occupational Displacement: A Multi-Regional Task Exposure Analysis of Emerging Labor Market Disruption

Authors: Ravish Gupta, Saket Kumar

Organizations: AI Lead, BigCommerce; IEEE Senior Member · University at Buffalo, The State University of New York, Buffalo, NY, USA

Abstract

This paper extends the Acemoglu-Restrepo task exposure framework to address the labor market effects of agentic artificial intelligence systems: autonomous AI agents capable of completing entire occupational workflows rather than discrete tasks. Unlike prior automation technologies that substitute for individual subtasks, agentic AI systems execute end-to-end workflows involving multi-step reasoning, tool invocation, and autonomous decision-making, substantially expanding occupational displacement risk beyond what existing task-level analyses capture. We introduce the Agentic Task Exposure (ATE) score, a composite measure computed algorithmically from O*NET task data using calibrated adoption parameters--not a regression estimate--incorporating AI capability scores, workflow coverage factors, and logistic adoption velocity. Applying the ATE framework across five major US technology regions (Seattle-Tacoma, San Francisco Bay Area, Austin, New York, and Boston) over a 2025-2030 horizon, we find that 93.2% of the 236 analyzed occupations across six information-intensive SOC groups (financial, legal, healthcare, healthcare support, sales, and administrative/clerical) cross the moderate-risk threshold (ATE >= 0.35) in Tier 1 regions by 2030, with credit analysts, judges, and sustainability specialists reaching ATE scores of 0.43-0.47. We simultaneously identify seventeen emerging occupational categories benefiting from reinstatement effects, concentrated in human-AI collaboration, AI governance, and domain-specific AI operations roles. Our findings carry implications for workforce transition policy, regional economic planning, and the temporal dynamics of labor market adjustment

Figures & tables

Panel of Figure 1 — Figure 1: ATE computation pipeline. O*NET task statements and ratings feed into task weights ( $w_{o,t}$ ), AI capability scores (CAP), and workflow coverage factors (COV). Regional adoption data parameterize the S-curve velocity function $V(t,r,\tau)$ . The four components are multiplied and summed across tasks to produce occupation-level ATE scores, which are then classified by displacement risk threshold.

Region	Tier	$k$	$\tau_{0}$	$L$	Calibration Basis
SF Bay Area	1	0.85	2024.25	0.92	Highest AI patent density and VC investment per capita [ 7 ]
Seattle	2	0.62	2025.00	0.85	AWS/Microsoft anchor employers; WARN Act data [ 16 ]
Austin	2	0.62	2025.00	0.85	Emerging hub; LinkedIn job posting trends [ 17 ]
Boston	2	0.62	2025.00	0.85	University-anchored AI cluster [ 7 ]
New York	3	0.48	2025.75	0.78	Diverse economy dilutes AI concentration

Table 1: S-Curve Adoption Parameters by Metro Tier

O*NET Ability	Category	AI Score	Benchmark Source
Written Comprehension	Cognitive	0.95	MMLU $\sim$ 87%
Written Expression	Cognitive	0.93	GPT-4 bar exam, CFA
Deductive Reasoning	Cognitive	0.88	MMLU Logic $\sim$ 88%
Information Ordering	Cognitive	0.85	AgentBench planning
Mathematical Reasoning	Cognitive	0.75	MATH benchmark $\sim$ 75%
Memorization	Cognitive	0.60	Context-window limited

Table 2: Representative CAP Ability-to-AI Mappings (Selected Abilities)

SOC Group	Workers (K)	Total %	Some Hrs %	All Hrs %	$\boldsymbol{R_{o}}$
Business & Financial Ops (13)	9,876	55.9	28.1	27.9	0.559
Legal (23)	1,822	52.7	33.2	19.5	0.527
Healthcare Practitioners (29)	9,974	13.1	8.1	5.0	0.131
Healthcare Support (31)	5,620	10.2	2.6	7.6	0.102
Sales & Related (41)	13,710	23.8	12.3	11.5	0.238
Office & Admin Support (43)	15,920	25.1	11.1	14.0	0.251

Table 3: BLS CPS Table 60: Telework Rates by Occupation Group (2024 Annual Averages)

	Employer Tier Share $\pi_{j}$			Mean ATE $\Delta$ (2027)
SOC Group	Tier 1	Tier 2	Tier 3	SF Bay	Tier 2	NY
Financial (13)	12.2%	27.1%	60.7%	$-$ 16.8%	$-$ 6.1%	$+$ 9.2%
Legal (23)	14.6%	23.2%	62.1%	$-$ 15.7%	$-$ 5.6%	$+$ 8.9%
HC Practitioners (29)	15.9%	26.7%	57.4%	$-$ 3.8%	$-$ 1.2%	$+$ 2.5%
HC Support (31)	15.9%	26.7%	57.4%	$-$ 2.9%	$-$ 0.9%	$+$ 1.9%
Sales (41)	15.9%	33.3%	50.7%	$-$ 6.5%	$-$ 1.8%	$+$ 5.0%

Table 4: Employer Tier Distribution

\pi_{j}

from QCEW 2024 and Remote Work Adjustment Impact (2027)

Panel of Figure 2 — Figure 2: Distribution of 2027 ATE scores across all 236 occupations for three representative regions. The dashed red line marks the moderate-risk threshold ( $\mathrm{ATE}\geq 0.35$ ). In the SF Bay Area (Tier 1), 70.8% of occupations exceed this threshold; in Seattle (Tier 2), 0%; in New York (Tier 3), 0%. The rightward shift from Tier 3 to Tier 1 reflects the adoption velocity differential parameterized by the S-curve model.

Occupation (SOC)	Category	2025	2027	2030
Credit Analysts (13-2041)	Financial	0.31	0.43	0.47
Judges & Magistrates (23-1023) ‡	Legal	0.31	0.43	0.47
Sustainability Specialists (13-1199)	Financial	0.31	0.43	0.47
Regulatory Affairs Specialists (13-1041.07)	Financial	0.31	0.43	0.47
Market Research Analysts (13-1161)	Financial	0.31	0.43	0.47
Preventive Medicine Physicians (29-1229) †	Healthcare	0.31	0.43	0.47

Table 5: Top 20 Occupations by Projected ATE Score (San Francisco Bay Area), 2027

	Seattle		SF Bay		Austin		New York		Boston
Category	’27	’30	’27	’30	’27	’30	’27	’30	’27	’30
Admin/Clerical (43)	0.0	74.5	78.4	92.2	0.0	74.5	0.0	0.0	0.0	74.5
Financial (13)	0.0	87.5	91.7	95.8	0.0	87.5	0.0	8.3	0.0	87.5
Legal (23)	0.0	100.0	100.0	100.0	0.0	100.0	0.0	14.3	0.0	100.0
Healthcare (29)	0.0	43.8	65.2	98.9	0.0	43.8	0.0	1.1	0.0	43.8
Sales (41)	0.0	63.6	68.2	95.5	0.0	63.6	0.0	0.0	0.0	63.6

Table 6: Share of Occupations Crossing ATE

\geq 0.35

(%) and Mean ATE by Region, 2027 vs. 2030

	Conservative			Baseline			Aggressive
Tier / Metro	2025	2027	2030	2025	2027	2030	2025	2027	2030
Tier 1 SF Bay Area	0.0	2.5	69.9	0.0	70.8	93.2	0.0	87.3	94.1
Tier 2 Seattle / Austin / Boston	0.0	0.0	3.4	0.0	0.0	60.2	0.0	19.9	70.3
Tier 3 New York	0.0	0.0	0.0	0.0	0.0	2.5	0.0	0.0	28.8

Table 7:

k

-Sensitivity Analysis: Percentage of Occupations Crossing

\mathrm{ATE}\geq 0.35

Risk Threshold. Three adoption-velocity scenarios stress-test the S-curve steepness parameter

k

, spanning a Conservative (

k_{\text{T1}}=0.40

) to Aggressive (

k_{\text{T1}}=1.20

) range. This range deliberately brackets the 6

\times

discrepancy between the calibrated Baseline value (

k=0.85

) and the empirically fitted trend from job-posting data (

k\approx 0.13

; Section 5.6 ), assessing whether the qualitative risk ordering SF Bay

>

Tier 2

>

New York holds regardless of

k

assumption.

AI and Labor

AI and labor studies how artificial intelligence changes jobs, wages, skills, hiring, and the organization and quality of work.

See all 95 papers

Who Delegates to AI? Evidence from Agent Configurations in Github
Aug 19, 2026Hyeongjae Lee, Jihyang Cheon, Lanu KimAgentic Workflows Organizational AI Adoption
Graduate School of Digital Humanities and Social Sciences, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea · Graduate School of Data Science, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea
Unaccountable Delegation, Fading Skills: Mapping the Risks of Workplace AI Agents
Aug 9, 2026Gabriele La Malfa, Lakmal Meegahapola, Edyta Bogucka +4AI Agent Safety Human-AI Collaboration
1King’s College London, UK · 2Nokia Bell Labs, Cambridge, UK · University of Sussex, UK +1
The Delegation Divide: An Open Index of AI-Agent Delegability across Occupations and Countries
Oct 22, 2026Cristian Espinal Maya, Santiago Jiménez LondoñoAI and Labor
Universidad Nacional de Colombia · Universidad EAFIT

Aug 19, 2026J/K move · Enter open · S save

Who Delegates to AI? Evidence from Agent Configurations in Github

Hyeongjae Lee, Jihyang Cheon, Lanu Kim

Graduate School of Digital Humanities and Social Sciences, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea · Graduate School of Data Science, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea

Agentic Workflows Organizational AI Adoption AI and Labor

Open paper AskPDF arXiv

A growing body of literature measures the extent to which occupations are exposed to AI, yet existing measures capture where AI could perform tasks rather than whether workers have actually adopted it. We introduce a distinct tier of exposure, delegated exposure, which records whether a worker has committed a task to AI by embedding it into a structured workflow. We operationalize this concept through the Agentic Adoption Index (AAI), measuring how closely an occupation's tasks align with the agentic routines that practitioners have built and shared. Using semantic embeddings of roughly 888,000 agent skill specifications from public GitHub repositories, we compute their similarity to nearly 18,000 O*NET task statements and aggregate these scores to the occupational level. We present three main findings. First, the occupations where task delegation concentrates differ sharply from those identified as most vulnerable by pre-AI automation frameworks. Second, the AAI aligns more closely with measures of technical capability than with measures of current conversational LLM use. Third, for occupations requiring a bachelor's degree or less, the AAI increases alongside average wage levels; however, this relationship reverses for occupations requiring a master's degree or higher, where adoption declines among higher earners. These patterns replicate on an independently collected corpus of agent skills from the Manus Skills Marketplace. This lower adoption among highly educated, high-earning workers may reflect tasks that inherently resist advance specification or professional discretion over the pacing of workflow codification. Distinguishing these mechanisms will require longitudinal measurement.

Abstract

AI and Labor

Explore similar work

Who Delegates to AI? Evidence from Agent Configurations in Github

Unaccountable Delegation, Fading Skills: Mapping the Risks of Workplace AI Agents

The Delegation Divide: An Open Index of AI-Agent Delegability across Occupations and Countries