When OpenAI quietly released ChatGPT as a research preview on 30 November 2022, it looked like a clever toy. Three years later, it has become something very different: a default starting point for questions, ideas, shopping – and increasingly, travel.
On its third birthday, ChatGPT is no longer just a chatbot. With hundreds of millions of weekly users worldwide and deep integrations into search, shopping, and booking platforms, it is evolving into an interface that sits between brands and the next generation of consumers.
That shift is being driven, more than any other group, by Gen Z.
From Viral Side Project to Everyday Infrastructure
In three years, ChatGPT has gone from a viral novelty to what many analysts now describe as “horizontal infrastructure” – a layer that can sit on top of almost any digital task.
Rough estimates put weekly active users in the range of 700–800 million. Its traffic growth has turned it into a significant referral source to publishers and creators, in some cases already ranking behind only Google and direct traffic as a driver of visits.
One columnist notes that ChatGPT is now the third-largest source of traffic to his Forbes content, with month-over-month growth strong enough to “create concern at Google”. That remark captures a broader anxiety in Silicon Valley: if ChatGPT becomes the default place to ask, search, compare, and decide, traditional search engines risk becoming the second click, not the first.
At the same time, the platform has shifted from pure Q&A to something closer to an operating environment. Users can now interact with third-party apps, browse the web, generate documents and code, and in some cases complete transactions without ever leaving the interface.
The first chapter of ChatGPT – experimentation, curiosity, and cultural fascination – is clearly over. The second chapter is about business models, commerce, and emotional connection.
Gen Z: The Normalisation Generation for AI
No cohort has embraced this new layer more quickly than Gen Z.
Young adults are the most intensive users of generative AI tools across workplaces, universities, and creative fields. Surveys consistently show that while older generations still treat AI as an optional add-on, Gen Z is folding it into everyday life: asking for restaurant ideas, building CVs, drafting essays, remixing content, and yes, designing holidays.
In travel, the generational divide is particularly stark. While a majority of boomers have never touched AI to plan a trip, a large share of Gen Z and millennials now use tools like ChatGPT at some point in the journey – from choosing destinations to sketching itineraries and comparing prices.
Yet this is not a simple story of enthusiasm.
Research out of major universities has found that many Gen Z students refuse to use ChatGPT not because of academic integrity concerns, but because of climate and ethical worries: the energy consumption of data centres, the dominance of a few tech giants, and the fear of outsourcing thinking itself. They are both the heaviest users and the loudest critics of generative AI.
In a sharply polarised political climate – what some commentators now call the “Divided States of America” – AI is an unusual point of convergence. Young people across the spectrum might disagree on policy, but they share similar habits when it comes to using AI tools to plan, book, and navigate experiences.
For brands trying to reach this demographic, the message is clear: Gen Z expects AI to be there, to be convenient, and increasingly, to understand them.
The Invisible Monetisation of GenAI
Against this behavioural backdrop, the economics of AI are starting to crystallise.
A new prediction from research firm Forrester, cited in recent commentary, suggests that by 2026 90% of consumers who use generative AI will help monetise it without even realising it. In other words, most people will think they are simply asking a question or planning a trip, while their behaviours quietly power new revenue streams.
That projection matches the strategic direction of AI platforms.
Subscription tiers such as ChatGPT Plus remain important, but they are unlikely to be the only pillar. The next phase is about turning conversational intent into purchases – and earning a slice of the transaction or advertising value along the way.
That might be a shopping assistant that recommends products in a chat. It might be a tightly integrated travel planner that can not only propose an itinerary but also book flights, hotels, and experiences through partner platforms. Or it might be context-sensitive promotions that surface in the middle of a natural conversation.
Forrester has also warned that a significant share of brands will damage customer trust with poorly designed genAI experiences. If nearly everyone is feeding data into AI systems, often unconsciously, the quality and transparency of those interactions will heavily influence how consumers judge companies.
In short, the second chapter of ChatGPT is not just about scale. It is about how seamlessly – or clumsily – those conversations turn into commerce.
From Defence to Offence: AI’s Travel Test Case
Most brands have so far used AI defensively.
In sectors from telecoms to banking, AI is already embedded in customer service, where it deflects calls, automates routine queries, and squeezes costs from complex operations. It speeds up search on websites, helps summarise documents, and optimises internal workflows.
Those efficiency plays matter. They will not disappear. But they represent only one side of what AI can do.
The more interesting frontier – especially for Gen Z – is offensive AI: systems that actively grow revenue by translating inspiration into action.
Travel is a natural test bed for this shift. It sits at the heart of the experience economy. It is rich in emotion, rich in choice, and rich in friction. The typical journey is full of drop-off points: a video on TikTok sparks a dream, but then reality intrudes. Flights need to be compared, hotel reviews read, prices tracked, and dates negotiated. Momentum fades.
AI promises to collapse those steps.
Imagine a single interaction that moves seamlessly from “I’d love to visit Japan next spring” to “Here is a three-city itinerary, aligned with cherry blossom season, with flights, hotels, and rail passes blocked in at your budget. Would you like to book now?”
That is the kind of offensive AI many travel brands are now racing to build – and it is precisely where ChatGPT is being positioned as a key discovery and orchestration layer.
HotelPlanner and Reservations.ai: Agentic AI in the Wild
One of the most aggressive moves in this space comes from HotelPlanner, a global travel technology firm that has quietly spent years preparing for a voice-first, AI-driven future.
At the Phocuswright conference in San Diego, co-founders John Prince and Tim Hentschel outlined their vision: agentic AI systems that don’t just answer questions but take action, acting like full-service travel agents at machine scale.
HotelPlanner has launched Reservations.ai, a travel technology suite that uses AI to plan, price, and book end-to-end trips. The platform is marketed as the engine behind a dramatic jump in HotelPlanner’s revenue over the past few years, and the company expects it to help push annual hotel sales above the US$2 billion mark.
Instead of static search boxes, Reservations.ai offers AI “reservationists” – named assistants that can handle both chat and voice conversations, switch across 27+ languages, and support customers in over 80 countries.
The experience is designed to mimic a conversation with a skilled human agent.
You might start with a vague desire: “I want to visit Japan.” The AI probes gently: Which month? What kind of trip is this – solo adventure, family, or honeymoon? Do you care more about price, convenience, or experience? Once you pick dates and cities, the system proposes a flight, waits for your feedback, adjusts, and then moves on to hotels and car rentals.
Each interaction becomes training data. Over time, the AI’s first suggestions are meant to “keep getting better”.
Voice, Context, and Emotional Intelligence
For HotelPlanner, the future of AI travel is built on three pillars: voice, context, and empathy.
“Eventually, voice-enabled travel will be a superior model,” Prince argues. Speaking is faster than clicking through multiple sites, and a well-trained AI can search pricing and availability much more efficiently than a human traveller juggling tabs.
But speed alone is not the breakthrough. Context is.
Traditional booking sites do not really know why a traveller is going somewhere. The same interface serves a parent taking a child to a football tournament, a couple on honeymoon, and a relative rushing to a funeral. The emotional and practical needs are entirely different, but the system treats them as identical input.
Voice AI changes that. You can say, out loud, what you are doing and why. “I’m travelling for a funeral, and I need something quiet, near the hospital, and not too expensive.” Or: “This is a once-in-a-lifetime honeymoon. Show me properties that feel special, even if they’re not the absolute cheapest.”
That “why” becomes part of the request, and therefore part of the response.
To respond appropriately, the AI must be trained not only on inventory and price data, but on emotional intelligence. Hentschel describes an effort to train models on millions of highly rated calls from human reservationists, rather than generic customer-service transcripts, to capture the nuance of tone, pacing, and empathy.
The goal, he says, is to bring empathy “so high that customers forget they’re talking to an AI”.
It is an ambitious claim, and one that will be tested in real-world deployment. But the ambition itself tells us where the battle is moving: no longer just on speed or cost, but on how “human” the AI feels without pretending to be a person.
Conversational UI: Voice Plus Visuals
Another shift on the horizon is the fusion of voice with rich visual interfaces.
HotelPlanner’s roadmap includes a conversational app where travellers can talk to an AI agent while simultaneously seeing photos and details of the rooms, pools, and local neighbourhoods being discussed. It is less like filling in a form and more like talking to an expert who controls a live, interactive brochure.
This vision closely aligns with the trajectory of platforms like ChatGPT, which are experimenting with multimodal experiences – blending text, voice, and images in a single thread.
For Gen Z, brought up on video-heavy social feeds and accustomed to speaking to their phones and smart speakers, this hybrid interface may feel more natural than legacy web booking flows. They expect to talk, swipe, and see at the same time.
If conversational UI does become the dominant way to book travel in the next decade, as Prince predicts, the platforms that combine voice, context, and visuals in a trustworthy way will have a powerful advantage.
From Sci-Fi Fear to Everyday Utility
The cultural narrative around AI has long been shaped by science fiction: rogue robots, runaway superintelligence, machines replacing humans completely. That imagery continues to influence public debates, especially around jobs and safety.
Yet on the ground, many of the systems gaining traction in 2025 look less like villains and more like tools – patient, consistent, and tightly constrained by design. Hentschel goes so far as to compare AI to a “friendly puppy”: endlessly available, not motivated by commission, and theoretically able to give honest suggestions based purely on user needs and data.
The reality, of course, is more complicated. Models are trained on human data, which carries biases and blind spots. They are controlled by profit-seeking companies, not neutral referees. And their environmental footprint is far from trivial.
But as they enter their third or fourth year of mass public use, a shift is visible: the fear-driven sci-fi narrative is giving way to a more mundane question – does this actually help me? For a Gen Z traveller trying to turn “I saw this beach on TikTok” into an affordable, safe, and memorable trip, that pragmatic test is what matters.
What 2026 and 2027 Could Look Like
Looking ahead, several trends seem likely to define the post-birthday era of ChatGPT and its peers:
- Commerce moves inside the chat: Travel, retail, and financial services will continue embedding their products into AI assistants, turning chat windows into storefronts and booking desks. The line between asking and buying will blur.
- Agentic AI becomes a competitive battleground: Simply answering questions will no longer be enough. Consumers will expect systems to take action on their behalf – within boundaries they can understand and control.
- Empathy and trust become differentiators: As Forrester warns, poor genAI implementations will erode confidence. Brands that invest in emotionally intelligent, transparent experiences will stand out, especially in emotionally charged domains like travel, healthcare, and education.
- Regulation and ethics move centre stage: With more people unknowingly helping to monetise AI through their everyday interactions, regulators are likely to intensify scrutiny around transparency, consent, and the use of behavioural data.
- Gen Z sets the bar: This generation will continue to normalise AI – in classrooms, workplaces, and airports – while insisting on higher ethical and environmental standards. Their expectations will ripple out to older demographics and to emerging markets.
A Birthday With Real Stakes
ChatGPT’s third birthday is not just a symbolic milestone for a single product. It marks the moment generative AI stops being a novelty and starts feeling like plumbing: invisible, always on, and deeply woven into how people search, plan, shop, and travel.
For brands, the question is no longer whether to use AI, but how – defensively to cut costs, offensively to grow revenue, or ideally both, without sacrificing trust.
For Gen Z, it is about turning an “answer engine” into an ally that can take them from inspiration to action, one conversation at a time.
And for the rest of us, it is a reminder that the interfaces we use today – chat windows, voice prompts, algorithmic suggestions – are quietly shaping the way the next chapter of the digital economy will be written.
