This post contains affiliate links. If you make a purchase through them, we may earn a small commission at no extra cost to you. As an Amazon Associate, we earn from qualifying purchases.
There is a property in Bangkok that looks, in the photographs, like exactly the kind of place you want to spend a week. Modern kitchen, large living room with the light coming in through floor-to-ceiling windows, a rooftop terrace with views over the city. The bathroom is the kind of minimalist Thai design that the expensive hotels aspire to. The price is competitive but not suspiciously cheap. The host has four reviews, all five stars, all warm.
The property does not exist. The photographs were generated by an AI that was given a prompt roughly equivalent to “luxury Bangkok apartment for Airbnb listing” and produced images of a space that is architecturally coherent, aesthetically pleasing, and entirely fictional. The four reviews were also produced by the same or similar AI, given a different prompt.
This is the specific problem that AI has introduced to the short-term rental market in the past two years, and it is getting harder rather than easier to detect as the technology improves. What was once identifiable by certain telltale distortions in generated images, the warped perspective, the wrong number of fingers on a decorative sculpture, the uncanny smoothness of every surface, is increasingly absent from the better AI image generators, which now produce photographs that pass casual scrutiny without difficulty.
How the Scam Works
The basic structure is simple enough. A fraudulent host creates a listing on Airbnb or a similar platform using AI-generated photographs of a property that either doesn’t exist or is not theirs to rent. The listing is priced attractively and located in a real area of a real city. The host’s profile may include AI-generated biographical text and a credible-looking photograph.
The initial booking process looks normal. Payment goes through the platform. The host communicates via the platform messaging system. The scam activates at one of several points:
The off-platform redirect. Before the stay, the host contacts the guest and asks to move communication to WhatsApp or email “for convenience.” They then request additional payment outside the Airbnb platform for a security deposit, cleaning fee, or service charge, typically to a bank account or through an unprotected payment method. This payment is not refundable through Airbnb’s dispute system because it happened outside the platform.
The arrival problem. The guest arrives at the address to find either that it does not exist, that it is a real address but is not available for rental, or that the property bears no resemblance to the photographs. The host is unreachable. The “apartment” turns out to be a shared hostel room, an unfurnished flat, or nothing at all.
The identity and payment phishing variant. The host contacts the guest before arrival claiming that “Airbnb requires additional identity verification” and provides a link to a fraudulent website that captures credit card details, passport information, or login credentials.
Why AI Makes This Worse
Prior to the widespread availability of AI image generation tools, creating a convincing fake listing required either stealing real photographs of a real property (which could be detected by reverse image search) or hiring a photographer to photograph a real space fraudulently. Both methods were detectable and both involved either technical risk (the original photographer or homeowner could flag the images) or logistical effort.
AI image generation removes both barriers. A convincing set of eight to twelve photographs for a luxury Bangkok apartment, interior spaces, terrace, bathroom, kitchen, the view from the window, can be generated in minutes with no physical property required and no photographs to reverse-search against a database.
The review problem has a similar structure. Genuine-looking five-star reviews written in the plausible style of satisfied guests, which is a stylistic task that large language models perform well, now appear on listings from hosts with no booking history.
Airbnb and similar platforms have deployed AI detection tools to identify generated images and anomalous review patterns. This works to varying degrees and at varying speeds. The detection systems improve, the generation systems improve, and the arms race continues in a way that the user should not assume is resolved in their favour.
How to Spot a Potentially Fraudulent Listing
Reverse image search. Run every property photograph through Google Images reverse search (right-click the image, “Search image with Google”) or TinEye. Stolen real photographs will appear connected to the original property; AI-generated photographs typically will not appear elsewhere but may reveal visual characteristics through the search process. This is not a perfect test, but it is a useful one.
Look for physical imperfections. Genuine interior photographs have real-world inconsistencies: slight misalignments, imperfect natural light, objects that are clearly in use rather than staged. AI-generated spaces tend towards an improbable perfection where every surface is pristine and every object is perfectly positioned. No lived-in quality. This is getting harder to use as AI improves, but it remains a partial indicator.
Check review patterns. Multiple five-star reviews posted within a short time period (within days or weeks of each other), particularly for a relatively new host, is a significant warning sign. Read the text of reviews carefully: AI-generated reviews tend to use similar sentence structures and similar phrases, and often describe the property in general terms that could apply to any similar listing rather than the specific details that genuine guests mention.
Host profile scrutiny. A host with no Superhost badge, with reviews that are very few in number and very recent, with a profile photograph that looks artificially perfect (no background, no candid quality), and with a listing history that is short, warrants additional caution.
Requests to communicate off-platform. If a host asks you to move communication to WhatsApp, email, or any channel outside the Airbnb platform before the stay begins, this is a primary red flag. Airbnb’s buyer protection is contingent on all communication and payment happening through the platform. Any request to move off-platform should be treated as a potential scam signal and reported to Airbnb.
Pricing. Listings priced significantly below comparable properties in the same area are worth additional scrutiny. This is not because discounted legitimate listings don’t exist, they do, but because the fraudulent listing is typically priced to generate bookings quickly.
What Airbnb Does and Doesn’t Protect You Against
Airbnb’s AirCover policy covers guests for specific problems including listings that are materially different from descriptions, properties that are unavailable at check-in, and hosts who provide false information. If you book through the platform, pay through the platform, and communicate through the platform, and the listing turns out to be fraudulent, Airbnb will generally refund your booking cost or find you alternative accommodation.
What AirCover does not cover: payments made outside the platform (the off-platform redirect scam), losses from identity theft, or bookings made on other platforms that use Airbnb’s name or branding fraudulently. This last category, phishing websites that mimic Airbnb’s design and capture payment details, is separate from the in-platform listing fraud described above.
Protecting Yourself: The Practical Checklist
Before booking any Airbnb listing, particularly for a high-cost stay of a week or more:
Reverse image search every property photograph. This takes five minutes and is the most reliable single check.
Read the reviews with attention to authenticity; do they describe specific details of the stay, or generic praise? Are there variations in writing style between different reviewers (indicating different people wrote them)?
Check the host’s profile: Superhost status requires a booking history and sustained positive reviews. A new host without Superhost status is not automatically suspicious, but deserves more scrutiny.
Book only through the official Airbnb app or website. If you find a property linked from a third-party site, go directly to airbnb.com to find and book the same listing rather than booking through the external link.
Pay only through the Airbnb payment system. Any request for additional payment through bank transfer, cryptocurrency, or external payment link should be declined and reported.
Save all communication. Screenshot conversations through the platform before your stay.
If you simply want to remove the authentication problem entirely, a hotel booked through Agoda or Trip.com resolves the problem at source. Hotel properties are physically verifiable, check-in involves a real reception desk rather than a key code sent by someone who may no longer be reachable, and the booking terms are standardised. For longer stays, serviced apartments booked through hotel platforms offer most of the space advantages of Airbnb with considerably more certainty about what you’ll actually find on arrival.
Consider booking with a credit card rather than a debit card. If a dispute arises, credit card chargeback protections provide an additional layer of recovery that debit cards do not.
After booking: contact the host with a specific question about the property; ask about a detail visible in the photographs, or ask about the procedure for collecting the key. A fraudulent host may not respond, may respond with a request for off-platform communication, or may provide an answer that is inconsistent with the listing.
The Booking.com and VRBO Equivalent
The same fraud exists across platforms. Booking.com, VRBO, and similar platforms all have fraudulent listings; the AI generation problem is not exclusive to Airbnb. The same protective measures apply across all platforms, and the fundamental principle, keep all communication and payment on the platform, scrutinise any request to move off it, holds equally.
If you want an alternative platform that tends to attract more established property managers, VRBO is worth bookmarking. It concentrates on whole-home rentals and skews toward hosts with longer booking histories. The same off-platform payment rules apply, but the listing pool is generally more conservative than Airbnb’s open-market model.
What to Do If You Are Defrauded
If you arrive at a property and it is materially different from the listing, unavailable, or non-existent:
Contact Airbnb immediately through the app. Document everything with photographs. Airbnb’s resolution process under AirCover is initiated through the app; beginning this process immediately is important.
If you have made a payment outside the platform, contact your bank or card provider immediately to report the fraud and initiate a chargeback if applicable.
Report the listing to Airbnb using the in-app reporting function. This removes the listing and flags the host’s account for investigation.
File a report with Action Fraud (UK) if you are a UK resident, or the equivalent fraud reporting authority in your country. This does not recover money but contributes to the data that law enforcement uses to pursue the fraudsters.
Frequently Asked Questions
How common are AI-generated fake Airbnb listings? Common enough to be a documented problem that Airbnb acknowledges; uncommon enough that most bookings with appropriate precautions go entirely smoothly. The risk is concentrated in high-demand markets (Bangkok, Bali, Barcelona, Lisbon, New York) and in listings priced to attract bookings quickly.
Is Airbnb still safe to use? Yes, with the precautions described above. Millions of Airbnb bookings complete successfully every week. The fraud exists at a fraction of the overall listing volume, but that fraction is large enough in absolute terms to require awareness.
Does AirCover protect me against AI fake listings? If the booking is made through the platform and all payment goes through the platform, and the listing turns out to be fraudulent on arrival, then AirCover should provide recourse. The protection is significantly reduced if you made any payment off-platform.
The AI generation landscape changes rapidly. Red flags that identify AI-generated images reliably in 2026 may not apply in 2027 as image generation technology improves. Stay current with booking platform security guidance and rely primarily on the off-platform communication and payment checks, which remain the most reliable fraud indicators.
A note on how I write: Every trip starts with a journal, filled in wherever I happen to be; on a bus, plane, train, or in a backstreet cafe that has clearly seen better days. It’s not tidy, and it’s not always legible even to me. When I get the time, I use AI to help turn those notes into something readable, organising drafts and occasionally rescuing a sentence that has wandered off without me. The experiences, the perspective, the humour and every editorial decision remain entirely mine. As for the photos, I’m not remotely fond of having my picture taken, family portraits included, so I also use AI to help visualise the places and moments I write about.
