The interface is dead. Welcome to the era of intent

SECTION 1

Introduction

OpenAI's latest Dev Day gave a tantalizing glimpse of the next phase of generative AI, not just new features, but the arrival of a new kind of digital experience. Instant Checkout, paired with Apps in ChatGPT, signals the emergence of a unified ecosystem where conversation, commerce, and creativity coexist.

The underlying ideas hint at the next era of digital products, where interfaces disappear, intent replaces navigation, and personalized interaction becomes the norm. These innovations aren't just about boosting an LLM's usefulness; they redefine what a digital channel can be.

SECTION 2

From interfaces to intent

Every new wave of interface has redrawn the digital map: the web browser, the smartphone, the voice assistant. You can trace the progression toward easier, richer interactions, and how that made business models more fluid. Apps in ChatGPT represent a new opportunity, one where conversation replaces navigation and intent replaces clicks.

What are Apps in ChatGPT?, Apps in ChatGPT are third-party or specialized tools that live inside the ChatGPT environment, extending what the model can do. In this new world, your users won't launch an app or type a search, they'll simply ask ChatGPT to plan and book their flights to Ibiza or create a Spotify playlist.

The initial launch partners (Booking.com, Expedia, Zillow, Canva, and Spotify) illustrate both the creative breadth of this ecosystem and its potential to solve some of the most frequently requested generative AI use cases, like trip planning and booking or personalized content generation.

The rise of AI-native interfaces, But the real headline is how this marks the emergence of AI-native interfaces: it creates an opportunity to meet customers where they already are, in natural-language conversations, rather than waiting for them to navigate to a website or app. It marks a profound shift in how discovery, intent, and purchase converge.

For enterprise architects, this raises both opportunities and challenges. Traditional UI flows are giving way to dialogue-based orchestration, where the model interprets context, invokes the right app, and executes a task seamlessly.

At Adservio, we see this shift as more than an incremental UX evolution. It's a fundamental break in how users interact with digital systems. For decades, we've asked users to adapt to our interfaces: learning where to click, which menus to open, which forms to fill in. The era of intent reverses that dynamic: it's the systems that adapt to users, understanding their intent expressed in natural language and orchestrating the necessary actions behind the scenes.

SECTION 3

Building an AI-native platform

For businesses, this is more than an integration opportunity. It's a fundamental shift in how you engage with your customers.

A new business channel, This isn't just a new interface: it's a new business channel. Embedding commerce inside AI assistants opens the door to experiences that are proactive, personalized, and context-aware. The moment recalls the early shift to Web 2.0, or the second generation of iOS apps, when digital products evolved from static experiences into platforms for deeper connection and ongoing engagement.

This way, businesses can embed services directly into conversational flows, from HR requests to IT automation or customer support. The result is a platform that unifies marketing, product, and service touchpoints into a single intelligent interface. Those who learn to design for this channel early will shape the playbooks for AI-driven customer experience over the next decade.

Concrete application examples, Imagine an employee asking, "How many vacation days do I have left, and when can I take them given project deadlines?" The AI assistant could: 1. Query the HR system for leave balances 2. Check the project management system for deadlines 3. Check the team calendar for planned absences 4. Suggest optimal windows 5. Initiate the leave request if the employee confirms

All of that within a single fluid conversation, without ever opening three different applications.

At Adservio, we help our clients design these conversational experiences by starting with identifying high-friction user journeys, those that currently require navigating between multiple systems or performing repetitive tasks. These journeys become the ideal candidates for conversational transformation.

SECTION 4

The money question

The seamless combination of conversational AI (OpenAI) and payment infrastructure, in this case, Stripe, fundamentally redefines people's buying experience while addressing a key point of friction for consumers: trusting services with their money.

### From click to intent

Instead of multiple clicks and checkout forms, transactions happen through dialogue: "I like that one, find it at the best price, and buy it." The model shifts from navigation-driven to intent-driven, reducing friction and making digital experiences more natural.

This transformation of buying isn't just a matter of convenience. It's a fundamental shift in the relationship between consumer and commerce. For years, conversion-funnel optimization has focused on reducing the number of clicks, simplifying forms, eliminating distractions. But even the most optimized checkout remains an experience break: you shift from browsing and discovery into a formalized transactional process.

With intentional commerce, that break disappears. Buying becomes a natural continuation of the conversation, not an interruption.

### The grey areas of monetization

But for retailers, this raises other questions: How do we make sure ChatGPT evaluates our offers fairly? How can we guarantee we're delivering the right products at the right price?

And so far, the biggest unknown is how monetization will work. We know OpenAI plans to introduce the Agentic Commerce Protocol (ACP) as the foundation for payments and revenue sharing, but specifics haven't been disclosed. Drawing on how other marketplaces have matured, we can speculate that early incentives may attract brands to build on the platform, with the economics evolving once a critical mass of apps is established.

Nine months on, some of that uncertainty has started to clear: OpenAI launched Instant Checkout with Etsy and then Shopify as its first partners, before shifting strategy in early 2026 toward payment experiences hosted directly by merchants' own apps within ChatGPT, rather than a single centralized checkout flow. The Agentic Commerce Protocol remains the shared technical foundation, but value-sharing keeps being negotiated merchant by merchant rather than under one fixed rule.

Critical strategic questions, This raises familiar strategic questions:

Where does value accumulate: with the platform or with the participant? Digital platform history teaches us it's not binary. The most successful marketplaces create value for both sides, but the distribution of that value shifts over time. Amazon started with thin margins for sellers to build critical mass, then gradually raised fees. Airbnb started by taking modest commissions, then introduced service fees for both hosts and guests.

How will fees, commissions, and token usage costs be structured? The token question is particularly interesting. Unlike traditional marketplaces where the cost of processing a transaction is relatively fixed, the cost of a conversational interaction can vary enormously. A simple query ("Buy the same detergent as last time") consumes far fewer tokens than a complex search ("Find me a birthday gift for my 12-year-old nephew who likes science but not video games, €50 budget").

Who pays for that variability? The merchant? The consumer? The platform?

Will the platform evolve toward a "walled garden" or an open ecosystem model? Historically, OpenAI has positioned itself as relatively open, but the introduction of commerce could change that dynamic. The more transactional value the platform captures, the stronger the temptation to close the ecosystem.

There's also uncertainty about who bears the ongoing usage costs, particularly token consumption. For now, Apps in ChatGPT are only available to premium subscribers; if that continues, it could limit audience reach or create new segmentation challenges for brands.

@cite:commerce-agentique-piloter-une-croissance-plus-intelligente

SECTION 5

What your teams should consider

It's still early to think about how this shift toward conversational UI will play out, but there are several obvious areas to start considering.

### Customer experience and data access

There's a huge opportunity for more personalized, contextual user experiences within an LLM-based interface. But it's unclear how much customer data will be shared with developers, will OpenAI keep most of the insights?

What about the privacy/analytics trade-off? This could be key to defining the platform's true business value. Businesses need to plan for data governance and understand the limitations around user visibility and behavioral insights.

At Adservio, we advise our clients to adopt a "data minimalism" approach: collect only what's strictly necessary to deliver value, and be fully transparent about its use. In a world where consumers interact through AI assistants, trust becomes even more critical since the direct relationship with the brand becomes more abstract.

### Developer experience and governance

The developer approval and review process isn't yet defined, we might expect it to resemble Apple's App Store-style models.

Do I need to extend my APIs to be ACP-compatible, i.e. product inventory, checkout, delegated payment? It's not yet clear how open or scalable the ecosystem will be for enterprise-grade deployments.

Key technical areas to evaluate include: - SDK ease of use and integration experience - Test pipeline and CI/CD compatibility - Security boundaries and required system access

Until clear patterns of tooling, governance, and QA emerge, large-scale enterprise adoption remains uncertain.

### Security and responsible use

New integration surfaces mean new threat vectors, from data leaks to malicious app behavior. Past incidents underscore how human error and weak governance can expose sensitive data.

Organizations must prepare proactive guardrails and monitor vetting processes once OpenAI defines its app approval model. Continuous security evaluation and compliance reviews will be essential to maintaining trust.

At Adservio, we recommend a layered approach to securing conversational integrations: 1. Strong authentication: Verify user identity before any sensitive action 2. Granular authorization: Precisely define what the assistant can and cannot do 3. Full audit trail: Log every action with context for investigation 4. Rate limiting: Prevent abuse or anomalous behavior 5. Human validation: Require confirmation for high-impact actions

@cite:les-dangers-de-l-agentwashing-ia-comment-les-agents-ia

SECTION 6

Reinventing the customer-brand relationship

Beyond technical and strategic considerations, the era of intent raises a fundamental question: how will brands build and maintain customer relationships when direct interaction becomes rare?

The abstraction paradox, In the traditional model, brands invest heavily in their interfaces, websites, apps, physical stores, because those are the touchpoints where the relationship is built. Every interaction is an opportunity to communicate brand values, create a memorable experience, build loyalty.

In a world where users primarily interact via AI assistants, these touchpoints largely disappear. The user says "Buy me some coffee" to ChatGPT, and ChatGPT orchestrates the transaction. The brand becomes invisible, abstracted behind the conversational interface.

How do brands differentiate themselves in this context? How do they build preference if users never really "visit" their brand?

From presence to preference, At Adservio, we believe the answer lies in a shift from "brand presence" to "algorithmic brand preference." Instead of optimizing for visibility and engagement on your own channels, you optimize to be chosen by the algorithms serving users.

This means: - Product excellence: If your product is objectively better, AI assistants will recommend it - Data transparency: Provide rich, structured, up-to-date product data that AIs can easily interpret - Verifiable reputation: Reviews, ratings, and trust signals that can be algorithmically assessed - Competitive pricing: AI assistants will automatically compare prices - Delivery experience: Speed, reliability, and delivery quality become key differentiators

In this new dynamic, the winning brands won't necessarily be the ones with the best websites or the most creative marketing campaigns. They'll be the ones that excel on the dimensions AI assistants evaluate when making recommendations.

SECTION 7

Conclusion: The time to act

There's testing to be done. But if you're curious about how your business can leverage these changes to integrate AI into your products, reimagine your digital channels, and build responsibly for the future, you're already ahead. This moment belongs to those ready to reimagine how technology meets human intent, turning conversation into connection and interaction into impact.

At Adservio, we support our clients through this transition into the era of intent. We start by mapping existing user journeys to identify where conversational interaction can create the most value. We rapidly prototype conversational experiences to test hypotheses and learn. We build the technical infrastructure needed to expose your systems to conversational interfaces securely and at scale.

And above all, we help our clients develop a long-term strategic vision to navigate this profound transformation. Because the era of intent isn't just a new marketing channel or a new customer touchpoint. It's a fundamental redefinition of how humans interact with technology, and by extension, with brands and services.

Companies that understand and embrace this shift now will be the ones that define the standards for customer experience over the next decade.

Note: The statements and opinions expressed in this article are those of the authors and do not necessarily reflect the positions of Adservio.

FAQ

Frequently asked questions