At risk / Concierge
SOC 39-6012.00

Concierge

VerdictHigher Risk — changing now
Signal 01 · structural exposure
92
/ 100

Concierge is one of the clearer at-risk findings in this batch: a structural score of 8 out of 100, nearly identical to Travel Agent's 12 (also assessed at risk on this site), reflects how much of the job's day-to-day workload — answering routine questions about wifi, checkout times, amenity hours, and generic local recommendations — overlaps with what a chatbot already handles well.

Signal 02 · real-world AI usage
High
band, not a number

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): Concierges scored 0.372 on AI applicability, notably 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): Concierges shows measurable Claude usage tied to this occupation's tasks, though at low absolute volume — treat as directional confirmation, not a second precise score.

What this is not
Neither reading measures whether any single concierge has lost work to AI. That data does not exist anywhere yet — both signals describe tasks, not headcount. This page is not a prediction that this job disappears.
How we calculated this →

What a concierge actually does

Concierges help hotel guests and building residents get things done that fall outside normal front-desk service: securing a reservation at a fully booked restaurant, tracking down hard-to-find tickets, arranging transportation, recommending an itinerary tailored to what a guest actually wants (not a generic top-ten list), and solving problems that come up mid-stay, from a lost passport to a medical need for a specific pharmacy at midnight.

The job depends on a real, current network of local contacts — restaurant hosts, ticket brokers, tour operators, drivers — built and maintained over time, plus the judgment to know which of several options actually fits a specific guest's budget, taste, and situation. High-end hotel concierges (often members of Les Clefs d'Or, an international professional guild) handle a wide range of requests daily, some routine and some genuinely unusual, and are expected to make it look effortless.

Concierges also field a steady stream of practical questions (directions, wifi, checkout logistics) alongside the higher-touch requests. Most concierges start in front-desk or guest-services roles and move up; larger hotels may require a hospitality management background for senior positions.

Why it reads this way

Concierge is one of the clearer at-risk findings in this batch: a structural score of 8 out of 100, nearly identical to Travel Agent's 12 (also assessed at risk on this site), reflects how much of the job's day-to-day workload — answering routine questions about wifi, checkout times, amenity hours, and generic local recommendations — overlaps with what a chatbot already handles well.

Hilton's Watson-powered robot concierge "Connie" and Marriott's AI chatbots on Messenger and Slack are real, in-market systems built specifically to field this kind of guest-facing query. Unlike some hospitality roles, no legal license stands between a hotel and an AI concierge system: Les Clefs d'Or, the profession's best-known credential, is a career guild membership, not a state-issued requirement to do the job.

What's left after routine volume shifts to AI — securing a same-night table through an actual relationship with a restaurant host, reading a vague request correctly, handling a genuine emergency — is real, human, and not something current AI does well, but it's a narrower slice of the historical role than the routine-query work that's already moving to bots first.

Skills this role draws on

Local Knowledge/NetworkingCustomer ServiceProblem-Solving Under PressureMultitaskingVendor Relationship ManagementDiscretionMultilingual Communication (often preferred)
Pay & demand
$27,000–$58,000
10th–90th percentile, USD/year
Average: $38,950/yr
Demand
Medium
Growth outlook
Stable
Projected growth
1-2% (2024-2034)

Range source: O*NET OnLine / BLS OEWS, 2025
Average source: O*NET OnLine / BLS Occupational Employment and Wage Statistics, Concierges (39-6012.00), 2025

Research verdict
Assessed as of 2026-08-10
At risk

Concierge's structuralScore of 8/100 is among the lowest in this batch, nearly identical to Travel Agent's 12 (also at risk on this site), and Microsoft's Copilot-usage data corroborates it directly: Concierges scored 0.372 on AI applicability, well above the cross-occupation mean and banded High. Real, named, in-market AI systems already perform a large share of the job's routine volume — Hilton's Watson-powered "Connie" and Marriott's AI chatbots on Messenger and Slack are built specifically to answer the amenity/hours/local-recommendation questions that make up much of a concierge's day. No legal license protects this role: Les Clefs d'Or is a professional guild membership, not a state-issued requirement. The genuine human-judgment tasks that remain — a same-night table secured through a personal relationship with a host, reading a vague request correctly, handling a real emergency — are real and still human, but a narrower slice of the historical role than the routine-query work already shifting to AI first.

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 (4)
Last updated: August 2026Source: O*NET OnLine / BLS OEWS, 2025

Data sources & methodology

Salary data: O*NET OnLine / BLS OEWS, 2025. Average figure sourced separately: O*NET OnLine / BLS Occupational Employment and Wage Statistics, Concierges (39-6012.00), 2025.

Task descriptions: Based on O*NET occupational analysis (39-6012.00).

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: 1-2% (2024-2034), based on BLS Occupational Outlook Handbook.

Learn more about our methodology