Customer Service Representative
Structural exposure score derived from 2 independent sources (OpenAI "GPTs are GPTs" (human_rating_beta) + AIOE (Felten, Raj & Seamans)), each normalized against its own full reference population and combined by simple average, then converted to a percentile.
In real usage data, AI shows up in this job more often than in most occupations.
Technical detail
Based on Microsoft's "Working with AI" study of real Copilot conversations mapped to O*NET tasks (arXiv 2507.07935): Customer Service Representatives scored 0.41 on AI applicability, more than two standard deviations above the cross-occupation mean (0.159, stdev 0.098) across all 785 SOC codes studied — banded here as High real-world AI usage relative to other occupations.
Corroborating signal from the Anthropic Economic Index (Claude.ai/API conversations, a different product and userbase than the Microsoft data above): this occupation shows observed exposure of 0.70, among the highest of the 756 occupations AEI tracks (dataset mean 0.077, 95th percentile 0.385). Unlike most occupations on this site, this is not a low-volume signal — treat it as directional confirmation of the Microsoft finding, not a second precise score.
What a customer service representative actually does
Customer service representatives are the front-line voice of a company on the phone, in live chat, and over email. They field questions about orders, billing, and account status; explain products, policies, and troubleshooting steps; and process returns, exchanges, and refunds.
A typical shift means working from a queue: taking inbound calls or chats one after another, pulling up a customer's account history, working through a script or knowledge base, and logging notes on every interaction before moving to the next one. Representatives track down the source of a complaint, calculate what a customer is owed, complete the paperwork or system entries needed to close a case, and sometimes recommend a renewal or upsell while they have someone on the line.
It's a role built around structured, repeatable back-and-forth: understand a defined problem, resolve it using approved procedures, record what happened, and move to the next case — often dozens of times in a single shift. That structure is what separates this work from occupations built around unpredictable physical tasks or improvised, face-to-face judgment calls.
Why it reads this way
Customer service representatives sit near the top of nearly every independent measure of AI exposure we could find. Microsoft researchers, analyzing millions of real Bing Copilot conversations mapped to O*NET tasks, scored this occupation 0. 41 on their AI applicability scale — well above the cross-occupation average of roughly 0. 16 — placing it in a small group of occupations where people are already routing a large share of their day-to-day tasks through AI tools.
Separately, the Anthropic Economic Index, which tracks what people actually ask Claude to help with, found this occupation among the top 1% of all 756 tracked occupations for measured AI usage. The underlying tasks explain why: much of the job is answering a written or spoken question by retrieving account information and following a defined resolution process, which is close to what conversational AI systems are built to do well.
BLS also projects employment in this occupation to hold flat or decline slightly through 2034, even as the volume of customer contacts companies handle keeps growing — a sign that some of that growth is already being absorbed by chatbots and self-service tools rather than new hires.
Skills this role draws on
Range source: U.S. Bureau of Labor Statistics OEWS wage data (May 2024), accessed via O*NET OnLine wage report, SOC 43-4051.00
Average source: BLS Occupational Employment and Wage Statistics, May 2025 (Annual Mean Wage)
Safer, skill-adjacent careers
These aren't generic "consider retraining" suggestions. Each one shares real skill or task overlap with customer service representative work, and each one scores meaningfully higher on our structural exposure scale, with the reasoning shown below.
Both roles are built around one-on-one client communication and problem-solving, but training requires hands-on physical coaching and real-time in-person adjustment that AI can't replicate — score 88 versus 14.
Both roles center on direct, real-time interaction with people trying to solve a problem or have a good experience, but a tour guide delivers that in person, leading a live group through changing conditions no chatbot replicates — a genuine physical-presence protection scripted call-center-style service doesn't have.
As of August 2026, Customer Service Representative scores 7/100 on structuralScore (Medium confidence) -- among the lowest on the site -- and clears no protective category-2 licensing wall or category-3 physical-task requirement. Real, named deployments corroborate the exposure: Klarna's AI assistant did work equivalent to roughly 700 agents in 2024 before a documented 2025 partial reversal, and Salesforce's CEO said in September 2025 that agentic AI helped cut support headcount from 9,000 to 5,000; BLS separately projects a 5% employment decline through 2034 and attributes it directly to task automation. A countervailing Gartner survey found most contact centers are pursuing workforce redesign rather than mass layoffs -- a genuine complicating signal, disclosed here, that does not change the occupation-level verdict.
This is a researched judgment (checklist v1.2), not a statistical measurement — it sits alongside, not instead of, the structural exposure signal above. Next scheduled review: 2027-02-10.
Sources (9)
- https://www.bls.gov/ooh/office-and-administrative-support/customer-service-representatives.htm
- https://digitaleconomy.stanford.edu/publication/canaries-in-the-coal-mine-six-facts-about-the-recent-employment-effects-of-artificial-intelligence/
- https://www.forbes.com/sites/quickerbettertech/2025/05/18/business-tech-news-klarna-reverses-on-ai-says-customers-like-talking-to-people/
- https://fortune.com/2025/09/02/salesforce-ceo-billionaire-marc-benioff-ai-agents-jobs-layoffs-customer-service-sales/
- https://www.cnbc.com/2025/09/02/salesforce-ceo-confirms-4000-layoffs-because-i-need-less-heads-with-ai.html
- https://www.gartner.com/en/newsroom/press-releases/2026-02-03-gartner-predicts-half-of-companies-that-cut-customer-service-staff-due-to-ai-will-rehire-by-2027
- https://www.gartner.com/en/newsroom/press-releases/2026-04-28-gartner-survey-finds-eighty-five-percent-of-service-and-support-leaders-are-expanding-human-agent-responsibilities-despite-expectations-of-mass-ai-layoffs
- https://www.careeronestop.org/Toolkit/Careers/Occupations/occupation-profile.aspx?keyword=Customer+Service+Representatives&location=UNITED+STATES&onetcode=43405100
- https://www.onetonline.org/link/summary/43-4051.00
Data sources & methodology
Salary data: U.S. Bureau of Labor Statistics OEWS wage data (May 2024), accessed via O*NET OnLine wage report, SOC 43-4051.00. Average figure sourced separately: BLS Occupational Employment and Wage Statistics, May 2025 (Annual Mean Wage).
Task descriptions: Based on O*NET occupational analysis (43-4051.00).
Real-world AI usage band: Microsoft's "Working with AI" study of Bing Copilot conversations mapped to O*NET tasks (arXiv 2507.07935), and corroborating data from the Anthropic Economic Index.
Growth projections: -5% (2024-2034), based on BLS Occupational Outlook Handbook.