> ## Content Index > Fetch the complete content index at: https://intelligence.artificialrealitymedia.com/llms.txt > Use this file to discover other available public pages before exploring further. # Five hotel jobs AI is already doing, and the one it keeps failing - URL: https://intelligence.artificialrealitymedia.com/hotel-jobs-ai-already-doing/ - Published: 2026-09-02T12:00:00.000Z - Updated: 2026-09-02T12:00:00.000Z - Description: Sofiane Ghorbel reports revenue management, forecasting, marketing, sentiment and operations running on AI in luxury hotels. The concierge desk is not. - Author: Galina Fendikevich - Tags: Hospitality, #For AI Buyers, #What's working, #Import 2026-09-03 13:11 Luxury hotels are already running AI in five places, and none of them is the front desk: revenue management, demand forecasting, marketing optimization, guest sentiment analysis and operational efficiency. Sofiane Ghorbel, founder of [AL Hospitality Group](https://www.alhospitalitygroup.com/?ref=intelligence.artificialrealitymedia.com) and a hospitality executive with more than twenty-five years across luxury hotels, resorts, travel, concierge services and destination management, reports all five in live use. The guest-facing concierge chatbot, which is where most of the industry's attention goes, is where it has been least successful. ## Which AI jobs are live in hotels today? Back-office ones. Asked where AI is genuinely deployed today, Ghorbel points at operations rather than reception. > "The hotel operators are already using AI ... for revenue management, for demand forecasting, marketing optimization, guest sentiment analysis, operational efficiencies." The outcome he attributes to it: > "The result is better service that they can provide, better profitability of course and better decision making." Note the order. Better service comes first. In Ghorbel's account back-office AI is not a cost-cutting story, it is what funds the service level. A property that forecasts demand accurately staffs correctly, and a guest experiences correct staffing as attentiveness. ## What does "technology creates efficiencies, humans create memories" mean? It is a resource-allocation rule, not a sentiment about being nice to guests. Ghorbel's position is that efficiency work should go to machines precisely so that the memory-making work, which is the actual product, gets more human attention rather than less. > "I believe AI will fundamentally transform our industry. However, AI will not replace hospitality. Hospitality is built on trust, judgment, relationships and human connection." > "The winners will not be the companies with the most technology. The winners will be the companies that combine high tech plus high touch." ## Which travel advisory tasks can AI take over? Seven, and all of them are preparatory. This is the workload Ghorbel names as compressible in his own field: - Destination research - Hotel comparisons - Itinerary creation - Market analysis - Translation - Client communication - Proposal preparation None of those is the judgment call at the end. That is what makes this a workable adoption model rather than a threat to the business: the tasks being handed over are the ones consuming the time that judgment needs. > "Let's say that AI is a force multiplier. AI allows small teams to operate like much larger organizations." > "AI does not replace expertise, it accelerates expertise." ## Can AI do concierge work? Half of it. Ghorbel grants that the capability overlap with the concierge role is real, then draws the boundary precisely. > "AI can now help remember preferences ... track special occasions, recommend experiences, identify patterns and anticipate requests. In many ways AI capabilities mirror the traditional concierge role." > "But the best concierge still provides something AI cannot: human judgment." He is not claiming machines cannot do concierge work. He is saying they can do most of the recall and pattern half, and none of the judgment half, and that in luxury the judgment half is what the client is paying for. ## Which AI products did he name? None. Across the interview Ghorbel does not name a single AI product, and that is a finding rather than a gap in the reporting. An executive with more than twenty-five years in luxury hospitality and specific views on where AI works can describe five live deployments without a vendor name attached to any of them. Read that as a market where the category is being adopted faster than any brand is being established. ## What to do about it this quarter Put the budget behind the desk before you put it in front of it. The five deployments Ghorbel reports as working all sit where hotel data already exists in volume. The guest-facing applications sit where it does not, which is covered in the companion article on why AI personalization fails for luxury clients. Then run the seven-task list against your own week. Anything on it that someone on your team is doing by hand is a candidate. Anything that ends in a recommendation to a client is not. Ghorbel puts part of the training obligation on the employer rather than the individual: "Even the company where they will work needs to have this will to train their staff how to use those many tools." He is blunt about the requirement for the people entering the industry, and optimistic about the result: "The role becomes more valuable and those professionals will spend less time processing information and more time creating relationship." The question worth answering before your next technology review: which hours on your team produce efficiency, and which produce memory? ## Where this comes from S1E3: Why AI keeps failing in luxury hospitality — the full interview with Sofiane Ghorbel. Listen or watch: [YouTube](https://www.youtube.com/watch?v=425kot6dd8I&ref=intelligence.artificialrealitymedia.com), [Spotify](https://open.spotify.com/episode/3SlI9LtC93dm3T52h1ZNLk?ref=intelligence.artificialrealitymedia.com) or [Apple Podcasts](https://podcasts.apple.com/us/podcast/why-ai-keeps-failing-in-luxury-hospitality-and/id6780699025?i=1000773200309&ref=intelligence.artificialrealitymedia.com).