AI agents: from chatbot sideshow to default transaction layer
AI agents in commerce are software systems that can understand natural language, reason over detailed data, and carry out multi-step actions such as planning trips, recommending products, or coordinating a home sale, turning fragmented online tasks into end‑to‑end automated transactions for consumers and businesses. AI travel agents have improved rapidly over the past two years, with KAYAK’s AI Mode in 2025 and its Ask AI expansion in 2026, while Booking.com keeps extending its trip‑planning tools and AI‑powered car rental assistance. At the same time, AI home‑selling platform Ridley was built to centralize and automate a process that has long been opaque and stressful for sellers. The through-line is clear: AI is no longer a gadget on the side of the customer journey; it is starting to run the journey itself.
This shift matters because the most painful consumer experiences are also the most data‑heavy. Travel itineraries, retail browsing, and property transactions all involve a mess of options, documents, and decisions. An AI that can see the whole picture can remove friction humans have come to accept as normal. But as these agents become the front door, they are also reordering power in commerce: whoever controls the data and the interface controls which brands even get seen.
Travel: AI agents chase the dream of one-conversation bookings
Travel is where the promise of AI agents is most intuitive: instead of wrestling with hundreds of tabs, you describe the trip you want and let the system do the work. AI travel agents now turn lengthy search processes into highly personalized shortlists, saving travelers time and surfacing options that better fit what they actually want. In practice, that means asking for a quiet hotel near a specific office with late check‑in and space for a large dog, and receiving a curated set of options instead of a wall of links.
KAYAK’s AI Mode and Ask AI, alongside Booking.com’s growing suite of trip‑planning tools and AI car rental search, push discovery toward natural‑language conversations rather than filter fiddling. These launches move the agent closer to completing the booking inside a single conversation instead of bouncing users through disconnected systems. Yet the industry’s Achilles’ heel is still data quality: agents struggle when hotels label a property as “pet‑friendly” or “accessible” without the granular details travelers rely on. Looking ahead, projects like Travala’s travel MCP, which lets AI agents search and book millions of hotels with programmatic payments, signal where this is going next. The winner in AI travel will not be the flashiest interface but the stack with the cleanest, richest underlying data.
Retail: everyone wants AI shopping recommendations, nobody wants to give up the data
If travel shows how AI can remove friction, retail shows how it can rearrange power. Shoppers are increasingly asking general AI systems for product recommendations, and retailers are racing to appear in those AI shopping results while resisting the pull to surrender the customer data that underpins online sales. Growing traffic from AI platforms has already pushed major chains to rewrite product descriptions and restructure sites so their items rank highly in chatbot searches. According to data cited in the reporting, 41% of consumers used generative AI for online shopping in June, and visitors referred by AI services generated 41% higher revenue per visit than those from traditional channels.
This is the new SEO: instead of optimizing for search pages, brands are optimizing for answers read aloud by an AI. Ulta, for example, reports doubled conversion and intent from shoppers finding its products through AI assistants. But retailers still want purchases to happen on their own sites, where they can track browsing habits, basket sizes, and past purchases to drive loyalty. The emerging pattern is grudging cooperation. Merchants need AI agents for discovery, but they fear becoming invisible suppliers inside someone else’s interface. Expect a tug‑of‑war over carts, loyalty programs, and checkout flows as AI shopping recommendations become a primary discovery channel.
Real estate: AI real estate automation goes after commissions, not chat
Real estate is where AI agents are attacking the business model itself. Selling a home involves pricing, promotion, legal paperwork, and negotiations that many sellers find exhausting. Ridley, an AI home‑selling platform, was created after its founder realized there was still no centralized system that helped consumers understand the process, coordinate the work, and make informed decisions from start to finish in the age of AI. The platform answers the question of whether AI can sell your home with a resounding yes, automating tasks traditionally handled by agents.
Crucially, Ridley is not pitched as a gimmicky chatbot. Its core “Ridley Intelligence” keeps persistent awareness of the transaction: property details, participants, contracts, deadlines, disclosures, inspection findings, communications, and what needs to happen next. Over the next five years, its CEO expects most coordination work—tracking deadlines, moving documents between parties, scheduling inspections, and following up with lenders and title companies—to disappear into software. That is AI real estate automation as AI commerce disruption: instead of taking a small efficiency slice, it goes after the entire coordination layer that justified hefty commissions. Traditional agents will survive only if they can prove their human advice is worth more than an always‑on transaction brain.
Support bots, faster shipping—and a growing privacy backlash
The same pattern shows up in travel accommodation. Airbnb rolled out an AI support agent in North America in 2025 and has since expanded it to more than 50 languages. The company says 45% of customer issues handled by this AI are resolved without human involvement, cutting support cost per booking by 16% year over year. Behind the scenes, AI now writes most of its code, shrinking the time from concept to feature launch by as much as 60% and helping ship nearly 80% more features in the first half of this year compared with the same period in 2024. This is AI customer support as infrastructure: not the flashy feature, but the reason new features appear faster and support queues feel shorter.
Yet all of these gains rely on a simple truth: better AI needs more customer data. Travel agents need detailed profiles and inventory; retailers want to keep browsing and purchase histories; real estate platforms depend on deep transaction records. At the same time, consumers and regulators are more protective of privacy, and retailers explicitly want purchases to stay on their own properties so they can retain data for future sales and loyalty. The next phase of AI commerce disruption will be defined less by clever agents and more by governance: consent frameworks, data‑sharing standards, and technical limits on what these systems can remember. The real question is not whether AI will run more of shopping, travel, and real estate—it already does—but who gets to see, store, and profit from the trails we leave behind.
