“Be More AI-literate” is Not a Skill. It is a Slogan.

If hospitality is to capture the value of artificial intelligence, its people need to know exactly what to learn — specific, teachable, assessable capabilities tied to the work they actually do. This paper sets out those capabilities role by role, because the AI skills a revenue manager needs share almost nothing with those a housekeeping supervisor needs, and a one-size curriculum serves neither.

The premise throughout is simple and now well-evidenced: value comes not from owning the cleverest model but from diffusing it — putting capable tools into the hands of trained, confident people. Two principles follow. First, skill is task-level: decompose each role, and the AI skills reveal themselves. Second, every technical skill below sits on top of, never instead of, hospitality’s human craft. The goal is a workforce that is fluent with the machine and irreplaceable without it.

00. How to read this guide

Each section names the specific skills for one department, written so they can be lifted straight into a training plan or a job description. Treat the three-point scale below as the proficiency target for every skill: most staff should reach Capable; team leads and power users should reach Fluent. A quick-reference matrix at the end summarizes the signature skills for each department.

A Shared Proficiency Scale

Aware. Understands what the tool does and where it fails; uses it with supervision.

Capable. Uses it independently for routine tasks and verifies every output before it reaches a guest or a ledger.

Fluent. Directs and orchestrates the tool, adapts prompts on the fly, knows when to override it, and coaches others.

01. The common foundation — every employee

Five skills no one in the building should be without, whatever their role.

• Practical prompting  —  giving an AI a clear instruction, context, and an example, then refining the result rather than accepting the first answer.

• Output verification  —  treating AI output as a draft: spotting plausible-sounding errors (“hallucinations”) and never sending unchecked content to a guest or into a system.

• Data & privacy hygiene  —  knowing which guest or commercial data may never be pasted into a tool, recognizing personal data, and using only employer-approved applications.

• Tool awareness  —  knowing which approved AI tools exist for their tasks, what each is for, and when not to reach for one at all.

• Knowing the limits  — recognizing the moment a situation needs a human — distress, complexity, sensitivity — and escalating rather than deferring to the machine.

02. Front Office & Guest services

The most guest-facing department, where AI fluency is felt directly in the experience.

• Co-writing guest communications  —  drafting confirmations, replies, and recommendations with AI, then editing for warmth and accuracy so the message sounds human and on-brand.

• Live translation fluency  —  using real-time translation to serve guests across languages while preserving tone and courtesy, not just literal meaning.

• Chatbot / AI-concierge hand-off  —  knowing when the automated assistant should escalate to them and taking over a conversation seamlessly, with full context.

• Knowledge-based prompting  —  querying the property’s AI assistant for policies, local tips, and availability to answer guests instantly and consistently.

• AI-assisted service recovery  —  using AI to draft a fast, fair response to a complaint while owning the empathy and the apology personally.

• Acting on next-best-offer prompts  —  interpreting AI-generated upsell and personalization cues at check-in without sounding scripted.

03. Reservations & Revenue Management

Where AI is most mature — and where blind trust is most expensive.

• Interpreting, not just trusting, recommendations  —  understanding the “why” behind an AI price or forecast and knowing when local knowledge should override it.

• Scenario prompting  —  asking the system structured “what-if” questions about events, demand shifts, and competitor moves to stress-test a strategy.

• Agentic distribution oversight  —  supervising automated rate and inventory updates across channels and catching anomalies before they propagate.

• Agent-to-agent literacy  —  recognizing that guests increasingly arrive through AI assistants that search, plan, and book on their behalf, and that the property’s own systems now transact with these external agents; treating the AI assistant as a distribution channel to be understood, not just a tool to supervise.

• Forecast hygiene  —  feeding clean, complete data, recognizing when a model has drifted from reality, and flagging it.

• Data storytelling  —  turning AI analytics into a clear commercial narrative for owners and general managers who want the decision, not the dashboard.

04. Sales & Marketing

The department most transformed by generative tools — and most exposed to brand risk.

• Generative content direction  —  briefing AI for copy, imagery, and video, then editing rigorously; raw, unedited output never ships.

• Answer-engine optimization (AEO/GEO)  —  structuring content so the property is surfaced and recommended by AI search and assistants, not only by traditional search engines.

• Brand-voice guardianship  —  maintaining a defined voice and a shared prompt library so AI output stays recognizably the brand’s, guarding against the “brand dilution” that generic AI text causes.

• Segmentation & persona prompting  —  using AI to draft audience segments and campaign variants, then validating them against real guest data.

• Campaign analytics interpretation  —  reading AI performance analysis and reallocating spend and attention accordingly.

• Responsible visual generation  —  creating and editing imagery with attention to authenticity, rights, and honest representation of the property.

05. Food & Beverage

Where AI quietly attacks the sector’s highest costs: waste, labor, and inconsistency.

• AI menu engineering  —  generating menu descriptions, analyzing item profitability, and tagging allergens and dietary information accurately.

• Covers forecasting for waste reduction  —  interpreting AI demand forecasts to plan prep and ordering, cutting food waste and stock-outs.

• Ordering & kiosk support  —  guiding guests through AI ordering systems and troubleshooting them without losing the hospitality.

• Inventory & substitution prompting  —  using AI to optimize par levels and suggest substitutions when supply changes.

• Responsible table personalization —  drawing on guest-preference data, within privacy rules, to make genuinely useful recommendations.

06. Housekeeping & facilities

Often overlooked in AI planning, yet rich in practical, physical-AI skills.

• Predictive-schedule literacy  —  working confidently with AI-optimized room-assignment and cleaning schedules, and knowing when to adjust them.

• Supervising mixed human/robot crews  —  coordinating cleaning and delivery robots, handling the exceptions they cannot, and keeping quality consistent.

• Maintenance-alert triage  —  interpreting AI and sensor maintenance alerts and prioritizing the genuine risks over the noise.

• Voice and photo AI reporting  —  logging issues hands-free by voice or image and checking that the AI captured them correctly.

• Acting on sustainability dashboards  —  responding to AI energy- and water-optimization prompts as part of daily routine.

07. Finance, Procurement & Administration

High-volume, rules-based work — ideal for AI, provided a human stays accountable.

• AI-assisted reconciliation & reporting  —  drafting reports and reconciling accounts with AI, then verifying every figure before sign-off.

• Anomaly & fraud interpretation  —  reading AI flags on transactions and chargebacks and judging which warrant investigation.

• Contract & document analysis  —  using AI to summarise and compare supplier contracts, with a human committing.

• Budget scenario prompting  — modeling budget and cash-flow scenarios quickly to support faster, better-informed decisions.

• Automation oversight  —  supervising automated back-office workflows and owning the exceptions and the audit trail.

08. Human Resources & People Development

HR both adopts AI and governs how the rest of the organization uses it.

• Responsible AI-assisted recruitment  —  using AI to support screening while staying alert to bias and never delegating the hiring decision itself.

• Personalized L&D design  —  building role-specific training with AI — including the AI-skills training in this guide.

• Skills-gap analytics  —  mapping the workforce’s current capabilities against future needs and planning reskilling accordingly.

• Employee AI-assistant management  —  running internal AI assistants for staff questions on policy, scheduling, and benefits.

• Acceptable-use policy literacy  —  writing, teaching, and enforcing clear, humane rules for how staff may and may not use AI at work.

09. Leadership & General Management

Leaders need fewer hands-on tool skills and far more judgment, governance, and vision.

• Diffusion leadership  —  pairing a clear top-down AI vision with the permission, tools, and training that drive genuine bottom-up adoption.

• Task-level role redesign  —  leading the reinvention of roles around human–AI collaboration — reframing jobs rather than simply cutting them.

• AI governance & risk ownership  —  taking accountability for data privacy, cybersecurity, vendor selection, and the avoidance of vendor lock-in, and for who answers when an AI gets it wrong.

• First-party data stewardship  —  treating clean, unified, first-party guest data not as a back-office asset but as the foundation of trust and discoverability in an agent-driven market: it is what allows external AI assistants to represent the property accurately, and the competitive moat that fragmented data quietly forfeits.

• Vendor & ROI evaluation  —  assessing AI suppliers and costs soberly, and measuring the return on adoption rather than buying on hype.

• Brand & ethics stewardship  —  protecting the brand from dilution and the guest from over-personalization, hallucinated service promises, and other avoidable harms.

• Change leadership  —  creating the psychological safety to learn — making clear that no one is fully an expert yet — and credentialing the skills people gain.

10. Putting it to work

Sequence the learning: the common foundation first, for everyone, then the role-specific skills, then the leadership layer that holds it together. Make the proficiency scale explicit, assess against it honestly, and credential what people learn so the skills travel with them and motivate the next cohort. Above all, leaders should model the behavior: adoption follows the example set at the top far more reliably than it follows a memo.

When done well, the result is not a workforce replaced by AI but one amplified by it — staff who orchestrate capable tools while delivering the warmth, judgment, and presence that remain hospitality’s reason for being. That conviction — that technology must serve and amplify the human person rather than displace them — now reaches well beyond the trade press: Pope Leo XIV’s 2026 encyclical Magnifica Humanitas places the dignity of human work in the age of AI among the defining moral questions of our time.

—  Quick-reference skills matrix

A Pertlink practical paper, and a companion to “Managers of Infinite Minds.” Prepared June 2026.

Made with the help of various AI tools, but with a HITL

© 2026 Pertlink Limited. Prepared by Terence Ronson. All rights reserved.