After trying Muse, I liquidated my Airbnb shares.

Discussions on how AI will transform the business world have never ceased in Silicon Valley and Wall Street. But as theory becomes reality, the first ones to feel the chill might be the internet platform giants we know best.

Recently, on the popular financial podcast *The Synopsis*, host Drew engaged in an in-depth conversation with senior independent stock analyst Mostly Borrowed Ideas (hereinafter referred to as MBI), who boasts over 150,000 followers.

MBI recounted that about ten days after Meta launched Muse, he downloaded the app and began testing it. This testing ultimately led him to make a pivotal investment decision—to liquidate his heavily weighted position in Airbnb stock and further increase his holdings in Meta.

During this conversation, Drew and MBI also delved into how Meta’s AI agent product, Muse, will disrupt the business models of internet aggregator platforms like Airbnb, Booking, Uber, DoorDash, and Amazon. A brand-new business era—“Proactive Commerce”—may be beginning to unfold.

Left: Drew, Right: Mostly Borrowed Ideas

Liquidating Airbnb After Trying Muse

As a heavy user of Airbnb (having spent over $7,000 this year) and a significant investor, MBI initially dismissed, like many others, the notion that AI would disrupt online travel agencies (OTAs). After all, booking accommodations requires viewing photos, reading reviews, and even chatting with hosts—needs that traditional text-based AI (like early versions of ChatGPT) simply could not fulfill.

But Muse changed his perspective.

MBI shared his experience: “I asked Muse to recommend five stays within a two-hour drive from my home, taking into account my past experiences on Airbnb. Shockingly, Muse not only automatically browsed my history but even analyzed the travel short videos (Reels) I’ve saved on Instagram to accurately grasp my preferences. It browsed and filtered in the background just like a real person, finally presenting me with options rich in images and reasoning.”

When MBI spotted a cottage and asked if he could book it directly, Muse told him within a minute: yes, and by bypassing Airbnb to book directly, the price would be 60% cheaper. Muse had already linked his credit card, requiring only a single confirmation click from MBI to complete the payment.

"In that moment, I realized Airbnb might be standing against the tide of the future," MBI admitted.

Host Drew raised a defense rooted in traditional business logic: “Airbnb’s biggest moat is ‘trust’ and ‘exclusive supply.’ As a consumer, I don’t dare transact directly with strangers on the internet; I need Airbnb as an intermediary to guarantee refunds and after-sales service. Furthermore, many hosts do not have their own standalone websites.”

MBI countered this: AI agents can easily establish a “trust” assessment by scraping authentic reviews across the web. More importantly, in the AI era, a host only needs to tell an AI, “Sync this listing to Booking and VRBO for me, and manage the calendar,” and the barrier of so-called “exclusive supply” will instantly dissolve. Hosts have long suffered under the “hegemony” of single platforms; they were previously hindered by high friction costs in managing multiple platforms, but AI eliminates that friction.

Airbnb vs. Booking: Which Has Stronger Risk Resistance?

After sharing his experience of selling Airbnb shares, the two turned their attention to another giant in the online travel platform space: Booking.

The curse of direct traffic: Airbnb currently enjoys up to 90% “organic/direct traffic,” which explains its high profit margins. However, in the age of AI agents, consumers may no longer open the Airbnb app directly but instead rely on AI to coordinate their itineraries. MBI predicts that Airbnb’s direct traffic will plummet significantly over the next 5 to 10 years.

The shift in customer acquisition costs: Currently, one-third of Booking’s traffic is purchased from Google (with an annual marketing budget exceeding $5 billion). For Booking, shifting the traffic entry point from Google to AI (such as Muse or ChatGPT) might merely mean paying “tolls” elsewhere, and the cost per acquisition (CAC) could even decrease.

Valuation cushion: Booking’s current valuation is only 10–11 times EBITDA, essentially being priced by the market as a “traditional conduit”; whereas Airbnb trades at a premium of 30 times. If both eventually become mere underlying infrastructure for AI, Airbnb’s valuation compression will be far more brutal.

Host Drew highlighted another defensive advantage for Booking—its loyalty program: “Over half of Booking’s room nights come from loyal members. The loyalty program offers points and free upgrades, which will incline consumers to stick with Booking even when comparing prices via AI.”

Yet, the two reached a consensus: AI will never make these OTA platforms “better.” At best, they will barely maintain the status quo; most likely, their long-term compound earnings growth rate will be compressed by AI.

Disrupting Amazon: When AI Takes Over Your “Shopping Cart”

If the low-frequency, high-ticket tourism industry takes the brunt of the impact first, what about high-frequency e-commerce and local services?

Take Amazon as an example. MBI pointed out that Amazon’s greatest crisis lies in its advertising business, which amounts to a staggering $76 billion. “In the past, when we opened Amazon, we often mindlessly bought the top three search results—that was the revenue logic of bid-based ads. But AI agents won’t click on ads. If you instruct your AI with ‘prioritize quality, then price at equal quality, and finally logistics speed,’ the AI will absolutely rationally seek the optimal solution across the entire web for you.”

MBI gave an example where he saw an ad for a T-shirt on Meta and asked Muse whether he should buy it. Rather than protecting Meta’s proprietary ads, Muse actually told him that the brand’s cost-performance ratio wasn’t high and recommended a better alternative.

Host Drew agreed and outlined Amazon’s path forward: “Amazon’s current app experience is actually quite poor; only the ‘search-to-purchase’ flow works smoothly. They must immediately deeply integrate native AI capabilities into their own app to defend the ‘user interface’ (UI). As long as users are still accustomed to opening Amazon first when they have shopping needs, and provided Amazon’s AI is sufficiently useful, combined with its robust logistics fulfillment network, it can hold its ground.”

Why Are Local Services (Uber and DoorDash) Temporarily Safe?

Compared to OTAs and e-commerce, both Drew and MBI believe that DoorDash (food delivery) and Uber (ride-hailing) will face impacts somewhat later.

MBI analyzed: “These two fall under high-frequency, low-ticket impulse consumption. When ordering food delivery, users often don’t know exactly what they want and need to ‘browse.’ More importantly, food delivery operates on a three-sided network (platform, merchants, riders). While AI can place orders with restaurants on behalf of the platform, it cannot replace riders delivering food to doorsteps. Currently, DoorDash makes only 50 cents per order, indicating they aren’t extracting super profits, which ironically becomes a moat.”

Host Drew believes ride-hailing (Uber) faces greater risks than food delivery: “If a food delivery goes wrong, users need customer service for refunds, giving the platform immense value in after-sales support. But ride-hailing is different. In the future, with autonomous vehicles (AVs) like Waymo and Tesla entering the space, plus AI agents instantly comparing prices across all platforms (Uber, Lyft, Waymo) within seconds, user loyalty to Uber will drop to almost zero, making the price war in the ride-hailing market extremely brutal.”

Embracing the Era of “Proactive Commerce”?

Towards the end of the episode, the two experts jointly introduced a new concept that will define business in the future—“Proactive Commerce.”

Traditional commerce falls into two categories: “intent-driven” (e.g., going to Amazon to search for a book) and “discovery-driven” (e.g., scrolling through Instagram and seeing an ad that sparks a purchase desire).

AI agents will bring a third model. MBI painted a chilling scenario: “Airbnb now offers a ‘book now, pay later’ feature (advance booking with deduction seven days before check-in). In the future, AI agents like Muse will scan your email, notice you booked a hotel but haven’t paid yet, and actively pop up to tell you: ‘Hey, I just noticed the hotel you booked. If you go to its official website to checkout directly right now, you’ll save 20%. Would you like me to cancel the Airbnb booking and rebook it for you?’”

No commands needed from the user. AI tirelessly works in the background around the clock to find better solutions and proactively suggests alternatives.

In this new era, the business logic of platforms that rely solely on information asymmetry and traffic monopolies to charge exorbitant “tolls” is being uprooted. For investors, it is time to reassess your “platform-type” stocks: in a tomorrow where AI possesses autonomous agency, will they remain indispensable infrastructure, or expensive intermediaries soon to be bypassed?

Join the official Coincamps community:

X: https://x.com/coincamps

Telegram: https://t.me/coin_camps