A Chatbot Told Them No Visa Was Needed. They Missed Their Flight.
In August 2025, Spanish content creators Mery Caldass and Alejandro Cid arrived at the airport for a romantic trip to Puerto Rico. Caldass had asked ChatGPT whether they needed a visa. The AI said no. Technically, that was correct — EU citizens don’t need a visa for stays under 90 days. But the chatbot failed to mention the ESTA form, a mandatory U.S. travel authorization that must be completed before departure. Airport staff denied them boarding. The TikTok video of Caldass in tears racked up over six million views.
This story went viral as a cautionary tale about trusting AI blindly. But for anyone running a tourism business, it tells a different story. The real problem wasn’t that a traveler used AI. The real problem is that most AI tools used in travel today are general-purpose systems with no connection to the actual booking pipeline, regulatory databases, or business logic behind travel operations.
That gap between what AI promises and what it actually delivers is where tourism businesses lose money, lose customers, and lose time.
Why Generic AI Breaks Down in Tourism Operations
The tourism industry runs on complexity. A single booking touches multiple systems: property management, CRM, payment processing, availability calendars, regulatory requirements, customer communication, and post-booking logistics. These aren’t isolated tasks. They’re interconnected workflows where one missed step cascades into real problems — a wrong date, a missed authorization form, a lost reservation.
General-purpose AI chatbots (the kind any traveler can access for free) were not built for this. They generate plausible-sounding answers from training data. They don’t check live availability. They don’t verify visa regulations in real time. They don’t update your CRM or trigger a confirmation email.
The market itself is moving fast. The global AI in tourism market was valued at approximately USD 3.37 billion in 2024, projected to reach USD 13.38 billion by 2030, growing at roughly 28.7% annually. About 80% of travelers globally say they’re open to using AI for planning and booking their trips. And AI-powered chatbots already handle approximately 80% of customer service interactions in the tourism sector.
But here’s the signal that matters most: according to Skift’s State of Travel 2025 report, only 2% of travelers are currently willing to give an AI tool full autonomy to make and modify bookings without human oversight. Trust is low. And when AI tools fail — like in the Caldass incident — that trust erodes further.
What Is an AI Agent in Tourism
An AI agent is software that can execute multi-step tasks autonomously by connecting to your existing business systems. Unlike a chatbot that answers questions, an AI agent takes action: it checks availability, updates booking records, sends confirmations, adjusts pricing, files reports, and handles exceptions — across multiple platforms, without human intervention at every step.
In the context of AI automation for tourism, an agent might receive a booking inquiry, check room or tour availability in the property management system, apply dynamic pricing rules, process the reservation, update the CRM, send a personalized confirmation email, and schedule a follow-up — all in one continuous workflow.
The difference between a chatbot and an AI agent is the difference between someone who can answer your questions and someone who can do the job.
The Real Cost of Not Automating: Observable Signals
Tourism businesses that still run on manual processes show consistent, measurable patterns:
Response lag. The average customer expects a response within minutes. Tour operators and small hotels relying on manual email or phone-based booking still take 4–12 hours to reply during peak season. By then, the customer has booked elsewhere.
Booking errors. Manual data entry across disconnected systems — a spreadsheet here, a channel manager there — produces duplicate bookings, wrong dates, and missed special requests. Each error costs staff time to fix and risks a negative review.
Pricing rigidity. Hotels and tour operators using static rate cards miss revenue optimization opportunities. AI-enhanced revenue management systems have shown revenue uplifts of up to 10–15% for hotels that adopt dynamic pricing.
Seasonal staffing pressure. Tourism is seasonal. Hiring and training temporary staff every peak season is expensive and inconsistent. Many tasks these staff handle — answering FAQs, processing standard bookings, sending reminders — are repetitive and automatable.
Reporting overhead. Destination management organizations and multi-property operators spend dozens of hours per month compiling reports from fragmented data sources. That time has direct cost and indirect cost (delayed decisions based on outdated numbers).
Solutions: How Tourism Businesses Are Deploying AI Agents Today
1. AI-Powered Booking and Inquiry Handling
Deploy a trained AI agent that connects to your booking engine, channel manager, and CRM. The agent handles incoming inquiries across channels (website chat, WhatsApp, email), checks live availability, answers questions about policies and amenities, and processes bookings — 24/7, in multiple languages.
This is the highest-impact, lowest-risk automation for most tourism businesses. Around 65% of travel industry executives identify chatbots and virtual assistants as the most impactful application of AI in their operations. And personalized AI-driven offers have been shown to increase repeat bookings by approximately 25%.
The key difference from a generic chatbot: the agent is connected to your actual systems, trained on your specific policies, and capable of completing transactions — not just suggesting them.
2. Dynamic Pricing Automation
AI-driven pricing tools analyze booking data, competitor rates, weather patterns, local events, and demand signals to adjust your rates in real time. This is already standard practice in airlines and is expanding rapidly across hotels, tour operators, and experience providers.
At Arival 360 in San Diego, Peek’s CMO demonstrated how their AI pricing tool allows operators to make up to 50 pricing changes in a single day — something impossible with manual processes.
For small and mid-sized operators, this means accessing the same revenue optimization capabilities that large OTAs have used for years, without needing a dedicated revenue manager.
3. Automated Guest Communication and Trip Management
AI agents can handle pre-arrival communication, check-in instructions, itinerary updates, real-time alerts (flight delays, weather changes), and post-stay follow-ups. When a flight delay occurs, a well-integrated agent can automatically update the hotel PMS, reschedule ground transport, and send a proactive notification to the guest — with no manual intervention.
4. Operational Reporting and Data Consolidation
For multi-property operators, DMOs, and tourism boards, AI agents can consolidate data from dozens of sources, generate customized reports for stakeholders and funders, track performance metrics, and maintain compliance — continuously, in the background.
A Google report estimates that AI can save an average of 175 hours per year per employee by automating repetitive tasks. In a tourism operation, that translates directly to staff hours redirected from data entry and report compilation toward guest experience and strategic planning.
5. Customer Service Escalation and Disruption Management
When weather, strikes, overbookings, or cancellations disrupt travel plans, AI agents can instantly reach out to all affected customers simultaneously — offering rebooking options, processing refunds, and answering follow-up questions. This is a scenario where manual processes simply cannot scale. A single disruption event can generate hundreds of service interactions within hours.
Navan’s GenAI agent, for example, handles business travel disruptions end-to-end: finding the next available option, checking company policy, rebooking, and notifying stakeholders — all before the meeting ends.
6. Custom AI Agents Built for Your Specific Business Logic
Off-the-shelf tools cover common use cases. But every tourism business has unique workflows, policies, pricing rules, partner integrations, and customer segments. A boutique hotel group in the Alps operates differently from a multi-day adventure tour operator in Southeast Asia.
This is where Lab51 works. We build custom AI agents tailored to your specific business processes — from booking automation and dynamic pricing to guest communication and operational reporting. No generic chatbot templates. No one-size-fits-all platforms.
We design, build, and deploy AI agents that connect to your existing systems and execute your actual workflows.
Would you like to discover opportunities for your business? Fill out the form, and our team will get in touch with you:
The ROI Question: What the Numbers Say
Organizations deploying AI agents report an average projected ROI of 171%, with 74% achieving returns within the first year. That figure deserves some skepticism — these are self-reported projections, and actual results vary by implementation quality. But directional data from real deployments is consistent:
| Metric | Reported Impact | Source |
| Revenue uplift (hotels, dynamic pricing) | Up to 10–15% | GlobeNewsWire, RaftLabs |
| Sales ROI increase | Up to 20% | RaftLabs |
| Customer service handle time reduction | 15–52% depending on complexity | Tourism AI Network |
| HR inquiry resolution time | Reduced by up to 80% | Tourism AI Network |
| Repeat bookings from personalized offers | ~25% increase | Gitnux |
| Employee time saved on repetitive tasks | ~175 hours/year | IA Tourisme |
The tourism businesses that invested in AI automation during 2024–2025 are already operating at different efficiency levels than those that did not. By the first half of 2025, approximately 45% of travel-related venture capital funding was going to AI-enabled startups — up from 10% in 2023.
As AI agent capabilities improve and integration costs decrease, the gap between early adopters and laggards widens. Businesses that prepare their data infrastructure, map their workflows, and deploy purpose-built agents in 2026 will lead their markets. Businesses that wait will spend 2027 rebuilding what their competitors already have.
The first step is simple: map your most time-consuming processes. Calculate what they actually cost. Identify which ones are repetitive, multi-step, and cross-system. Those are your automation candidates.
That’s what we build at Lab51. If you’re ready to automate the right way, start a conversation at lab51.io.