For the past decade, the mechanics of advertising have remained fairly consistent. Brands compete for attention across search results, social feeds, video and retail media—optimizing for clicks, impressions, and conversion rates. While the surfaces have evolved, the basic model hasn’t changed: people navigate, platforms intermediate, and advertising monetizes the journey.
Artificial intelligence is changing the equation.
Instead of static messages placed around content, we’re entering an era where the interface itself understands intent, synthesizes options, and introduces commercial choices directly inside a conversation. This isn’t just a new marketing channel—it’s a fundamentally new advertising environment, with new mechanics, incentives, and rules of competition. Marketing organizations must be prepared to rapidly test, learn, and—most importantly—scale what works across this growing ecosystem.
From Neutral Copilots to Commercial Systems
In the last two years, large language models (LLMs) have become central to how people search, learn, and shop. Adoption has accelerated rapidly: a recent BCG study showed shopping related GenAI use grew by 35% in 2025. What started as a tool for casual assistance is quickly becoming the default layer of digital discovery.
As usage concentrates around a handful of platforms, business model challenges are predictable. Consumer-scale AI infrastructure is expensive. Subscriptions and enterprise licensing alone can’t (and won’t) sustain these platforms for long. Just like search and social, advertising is emerging as a likely monetization layer—transforming LLMs from neutral copilots into commercial systems.
Marketers are already reacting. Conversational advertising is emerging as a distinct line item in the media mix—separate from search and programmatic—as brands reserve dollars to test, learn, and build early capability. Today, 53% of organizations are allocating budget to this emerging format, with nearly three-quarters planning significant increases over the next two years (spend will likely to be reallocated more fluidly across the existing channel mix once performance benchmarks stabilize and integration improves).
The strategic question is no longer if advertising will appear in AI-driven experiences—but how: how ads are surfaced, how they are governed, and how brands compete and scale winning tactics in this new environment.
The AI Attention Stack Is Taking Shape
Consumers are spending less time navigating the web through predictable funnels of links, feeds, and category pages. Increasingly, they begin (and often finish) their journeys inside AI interfaces that combine exploration, evaluation, and recommendation into a single, cohesive experience.
This creates a new “AI attention stack”: the surfaces where time, intent, and influence concentrate—and where ad inventory will naturally form.
Three types of interfaces define this stack:
- Search-Embedded AI: These systems reshape discovery by collapsing multiple sources into a single synthesized response (e.g., Google AI Overviews, Perplexity, and Microsoft Copilot).
- Assistant-Native AI: Always-on problem solvers embedded into daily workflows—planning, researching, deciding, and suggesting (e.g., ChatGPT, Gemini, Claude, and Meta AI).
- Retail and Commerce AI: Specialized agents trained on retailer-owned data that shift shopping from product grids to goal-based conversation (e.g., Amazon Rufus, Walmart Sparky, and Instacart Ask).
Across these environments, advertising inventory is emerging in three distinct forms:
- In-answer ads, embedded directly within synthesized responses when commercial intent is detected.
- In-conversation ads, appearing alongside dialogue during exploration—even for non-transactional queries.
- Agentic ads, where sponsored options surface as an AI agent plans, compares, or completes tasks on a user’s behalf.
Each layer introduces new formats, rules, and competitive dynamics. Winning inside AI-driven discovery requires understanding not just where attention lives, but also how models interpret and elevate commercial relevance.
Discovery, Reinvented
Search is undergoing its most fundamental redesign since the introduction of paid listings. Generative systems are replacing ranked links with synthesized answers, and advertising is moving upstream into those outputs.
Google’s AI Overviews are an early signal. For exploratory and research-oriented queries, Shopping and Search ads now appear directly inside AI-generated responses, often ahead of organic results. Visibility now relies less on keyword coverage and more on how well product data, creative, and bids align with model-level interpretation.
Microsoft Copilot is following a similar path, reframing search as an ongoing conversation built atop Bing’s existing ad ecosystem. Perplexity—a popular AI-powered search engine—is testing a more explicit approach , surfacing sponsored follow-up questions that are clearly labeled and contextually generated—reflecting rising expectations for transparency in AI-generated experiences.
The exact mechanics may vary, but the direction is consistent: discovery is becoming synthesized, conversational, and increasingly commercial.
When Assistants Become Advertising Surfaces
General-purpose AI assistants are quickly embedding themselves into everyday workflows—from planning and research to execution and purchase. As they capture more attention, they are becoming the next major advertising surface, even as monetization models remain in flux.
“This is the first time in advertising history where brands can talk to customers one-to-one at scale—and actually listen to what comes back.” — Andrea Tortella, Thrad.AI
Today, most assistants don’t display ads inside conversations. But the infrastructure is quickly forming. OpenAI recently announced it will begin testing ads within ChatGPT for its U.S. users in the coming weeks. Meta already uses interactions with Meta AI to inform ad relevance across its platforms—blurring the line between assistance and targeting.
Let’s be clear—assistant-native AI advertising will not look like search advertising. It needs to feel native to conversation: suggested next steps, recommended tools, sponsored follow-ups, “best option” summaries, or embedded handoffs into transactional flows.
The limiting factor? Trust. A recent study revealed 69% of consumers feel manipulated when brands use AI for advertising without disclosing it. When users treat assistants as advisors, any perceived manipulation carries a higher penalty than an ad in a feed. The winners will be the platforms—and brands—that preserve credibility through clear disclosure, relevance, and restraint.
Retail Media’s Next Frontier
Retail media is reaching an inflection point. GenAI is shifting e-commerce away from retailer-owned destinations toward agentic search intermediaries that increasingly resolve intent upstream—reducing the volume of traffic that ultimately reaches retailer sites.
In response, retailers are launching proprietary shopping agents trained on their rich first-party data, and monetization has been quick to follow. Walmart recently announced advertising inside their shopping companion Sparky (following Amazon’s introduction of ads within Rufus in 2024), enabling sponsored products to surface naturally within conversational shopping flows. However, emerging standards like Google’s Universal Commerce Protocol (UCP) point to a broader structural shift, allowing retailer data, assortment, and fulfillment capabilities to be accessed—and monetized—outside owned environments.
The implication is a redefinition of retail media economics. For retailers, value shifts from monetizing on-site attention to participating in distributed, agent-led commerce. For brands, advantage moves away from grid placement and traffic capture toward being selected by models—driven by data quality, availability, and relevance—as AI systems dynamically assemble shopping outcomes across ecosystems.
A New Interface for Consideration
AI-driven interfaces may prove most transformative for categories where decisions require research, explanation, and trust—not just exposure. While digital advertising has long been effective for transactional products, many B2B and high-consideration B2C categories—such as financial services, insurance, healthcare, and enterprise software—have struggled to tell a complete story within feeds, banners, or keyword-driven formats.
Conversational AI changes that. These interfaces allow users to explore complex questions in one place—asking follow-ups, weighing trade-offs, and comparing options within a trusted, context-aware environment. When advertising appears in these spaces, it can function less as an interruption and more as guided decision support. For categories built on understanding rather than impulse, AI assistants offer a new way to participate meaningfully in the consideration process itself.
As these interfaces take on a more central role in complex decisions, the way advertising appears within them—and the standards governing transparency, data use, and trust—becomes increasingly consequential.
The New Rules of Advertising Inside AI
It’s clear that embedding advertising into AI-generated answers raises new expectations around transparency, data use, and neutrality. Consumers are already drawing lines. Nearly 70% identify certain
data categories
—private messages, health information, and precise locations—as off-limits for AI use. That sensitivity elevates risk across three dimensions:
- Content and publisher integrity, as models summarize third-party material while embedding paid influence.
- Data and targeting governance, particularly when conversational inputs inform broader ad ecosystems.
- Algorithmic neutrality, as organic and sponsored recommendations converge.
Regulatory momentum is building. Early proposals point toward clearer disclosure, stricter labeling, and tighter limits on conversational data usage. Platforms and brands that lead with transparency will be better positioned as guardrails harden.
Four Plausible Futures for AI Advertising
Despite rapid changes on the horizon, the trajectory isn’t fixed. Four scenarios illustrate where LLM-driven advertising could land:
- Search 2.0: Ads embedded directly into synthesized answers, optimized around intent clusters rather than keywords.
- Agentic commerce: AI agents manage research through purchase, shifting retail media toward influencing agent defaults.
- Ambient promotion: Commercial influence becomes algorithmic and contextual, without discrete ad units.
- Regulated neutrality: Strong guardrails constrain targeting and optimization in favor of clearer separation and disclosure.
The reality will likely blend all four. That makes preparation—not prediction—the priority.
What This Means for Marketing Leaders
Three implications stand out:
- Discovery Becomes Model-Mediated: Models evaluate relevance holistically. Winning requires product data quality, structured content, and creative designed for model comprehension—not just ranking mechanics. It also demands tighter integration across brand, performance, and retail media—because models merge content and placement in ways organizations often don’t.
- Retail Media Becomes More Complex: Retail media is extending beyond retailer-owned surfaces, likely to be distributed via standardized agentic commerce protocols. For brands, influence now travels with the product across ecosystems. Success depends on being agent-ready: clean SKU data, interoperable content, and the ability to adapt as recommendations are dynamically assembled.
- Governance Becomes a Differentiator: Trust will shape adoption. Leaders need clear internal rules for conversational data use, disclosure standards, and consent—before regulation makes those decisions for them.
How to Act
Things are moving quickly. The shift from feeds and links to answers and agents is accelerating unevenly, but decisively. The next 18 months are pivotal.
Marketing leaders should take these actions immediately:
- Get Access, Start Experimenting, and Scale What Works: Secure early access to alpha and beta programs across emerging AI platforms and move quickly into structured experimentation. Test placements within AI overviews, sponsored answers, and retail assistants to understand how models evaluate product data, category signals, and creative inputs. Apply these learnings to rapidly scale high-performing approaches and ensure your organization stays ahead as the ecosystem matures.
- Build the Operating System for AI-Native Media: Improve data hygiene, unify creative and media workflows, and adopt rapid experimentation. AI interfaces collapse silos—so organizations that remain fragmented will move too slowly.
- Establish Governance for Conversational Data and Disclosure: Define permissible data, labeling standards, and consent capture. In AI interfaces, trust is not a soft metric. It’s the cost of entry.
- Start building toward your Marketing Organization of the Future: Begin rethinking how your media mix, go-to-market model, and operating structure will need to evolve. As GenAI collapses planning, creative, and activation, marketing organizations of the future will be more integrated, more internalized, and more dynamically allocated.
The Future Isn’t Set—Yet
What comes next will depend on how platforms balance monetization with credibility, how brands adapt to being interpreted by models, and how consumers respond as conversations become commercial environments.
The shift is already underway. The next era of advertising won’t be defined by where ads appear—it will be defined by how decisions are shaped inside AI systems.
The organizations that prepare now won’t just adapt to the new environment. They’ll help shape it.
