AI Update – May 29, 2026: What Changed in Ads, Search, SEO, and Agentic Commerce

The past week was pretty intense in the world of Artificial Intelligence, with direct impacts on Advertising, SEO, search experience, and digital commerce. It is not an exaggeration to say that the AI layer is starting to rewrite the rules of the game for both brands and platforms.

Instead of just improving recommendations or completing text, AI is becoming an active intermediary between people, content, and products. Conversational assistants are starting to suggest brands, compare prices, organize reviews, and in some cases even make purchase decisions on behalf of the user. This shift is being driven by three main forces:

• Platforms like ChatGPT and Google redefining how ads and results show up;
• The rise of agentic commerce, with AI agents acting as buyers;
• Growing pressure for brands to become machine-readable through structured data and well-managed metadata.

At the same time, it is becoming clear that simply stacking automation is not enough. Consumers are starting to push back on campaigns that feel empty and robotic, companies are questioning the return on AI investments, and search engines are facing resistance to the forced adoption of experiences filled with model-generated answers.

ChatGPT as an Ad Platform and E-commerce Showcase

One of the most relevant moves is the expansion of Advertising formats inside ChatGPT. In recent tests, the system started supporting much more visual and transactional creatives, including:

• larger, richer images with strong visual appeal;
• customizable call-to-action buttons tailored to the context of the conversation;
• e-commerce-specific layouts, with price, customer reviews, and product information in the spotlight.

Behind the scenes, OpenAI has been building a much more robust ad infrastructure, with audience targeting, outcome-based optimization, conversion tracking, and self-serve campaign management. Interestingly, company executives have highlighted how creative performance varies a lot depending on the conversation context, which makes the chat environment very different from a traditional feed.

In practice, the platform is racing to turn ChatGPT into a scalable media channel that looks like a search engine, consultant, and marketplace all at once. All this is happening in parallel with market moves that suggest a possible IPO as early as 2026, which helps explain the rush to build a solid revenue model based on ads and commerce.

For people working in marketing, the message is clear: ads in conversational AI environments are not just recycled versions of banners or search campaigns. They require their own strategies for creative, measurement, and targeting, designed around dialogue, user intent, and the model’s dynamic response.

Agentic Commerce: When the Customer Is an AI Agent

Another key front is the rise of agentic commerce, where AI agents take over chunks of the research, comparison, and purchase decision process. Instead of users manually browsing through dozens of tabs, the agent analyzes stock data, return policies, reputation, price history, loyalty programs, and other customers’ reviews.

The direct consequence is that brand trust stops being only an emotional concept and becomes something measurable and machine-readable. In agent-driven scenarios:

• data inconsistency or lack of structured information may cause a brand to simply vanish from recommendations before the consumer even sees the option;
• operational data matters as much as communication: on-time delivery rate, clarity of refund policies, price consistency across channels;
• any noise in product metadata can tank visibility in systems that prioritize reliability.

Recent reports show that major retailers are already going hard in this direction. Surveys indicate that 44% of large e-commerce players are already integrating buying-agent protocols, and a significant share expects that up to 30% of transactions will go through AI agents within a few years. At the same time, less than a third of these companies say they are truly prepared for fraud and security risks tied to this new flow.

For marketing and product teams, this changes the foundation of the job: it is not enough to convince the human buyer, you also need to be convincing, verifiable, and trustworthy for the algorithms that filter the path to the cart.

Structured Data, Metadata, and the New SEO Foundation

In this context, structured data and metadata stop being a back-end technical detail and become a strategic asset. Search platforms, recommendation systems, AI assistants, and buying agents work much better when they can clearly understand what type of content or product they are analyzing.

Some of the elements gaining importance include:

• well-implemented schema markup for products, articles, events, services, and help content;
• complete product attributes: variation, size, color, compatibility, price, stock, and warranty;
• rich metadata for images, videos, and brand assets, helping models classify and recommend better;
• provenance and trust signals, especially in an environment full of synthetic content.

Platforms like Pinterest and Adobe have been using metadata heavily for personalization and asset discovery, and the same logic is spreading across search engines and large language models. In generative AI environments, this metadata becomes the foundation for understanding, ranking, and activating content.

In practice, the role of anyone managing SEO goes way beyond keywords. You need to act as a bridge between human content and machine reading, making sure everything the brand promises to people is mirrored in structures that models can decode without errors.

Google, AI in Search, and the Fight for Visibility

Google has also been speeding up important changes in how it delivers AI-powered results. The company has been expanding AI Overviews and AI modes, consolidating automatically generated answers ahead of traditional links. To reduce the feeling that publishers are being sidelined, a few new features have appeared:

• user-selected favorite sites, which get a preferred source badge inside AI answers;
• highly cited labels highlighting original reporting and content with strong cross-referencing;
• carousels of articles and content from sources considered trustworthy.

Internal company data suggests that links with the favorite badge receive almost twice as many clicks compared to regular results. At the same time, the architecture of paid media campaigns is also changing. To show up well in the new AI search experiences, Shopping, Performance Max, and broader automation-based campaign types are becoming central.

Instead of relying so heavily on exact keyword match, the system now factors in:

• product feed quality;
• consistency between landing page and search intent;
• correct structured markup on the site;
• creatives written in a more conversational, informative tone.

In parallel, the SEO community has been debating the idea that brand strength matters more than ranking for a single query. With search engines, answer engines, and tools like ChatGPT, Reddit, LinkedIn, and YouTube fragmenting how people discover information, the thesis that YBYS — Your Brand = Your SEO is starting to gain traction.

For marketing professionals, this means that reputation, recurring exposure, community presence, and credible mentions tend to sustain long-term visibility, even in a scenario where users receive ready-made answers without necessarily clicking.

User Pushback and the Fragmentation of Search

Not everyone is excited about having AI in everything. The negative reaction to changes on Google’s results page, with long AI-generated answers taking over the main area, is already starting to show up in the numbers.

DuckDuckGo, which positions itself around privacy and control, reported a sharp increase in app installs and search usage after Google announced that AI agents would take on a more central role in the default experience. In the US, the search engine saw weekly install spikes above 30% for its apps, especially among users looking for AI-optional experiences instead of AI forced by default.

DuckDuckGo’s model bets on search modes without generative AI turned on and offers AI tools as an add-on, activated by user choice. This positioning resonates directly with people who are uncomfortable with automatic summaries, image generation, and heavy collection of contextual data.

In practice, this resistance breaks the idea of universal, homogeneous AI adoption in search. For brands, it means planning a presence across multiple environments: from AI-heavy experiences to more traditional search engines and privacy-focused spaces.

Programmatic Advertising and AI-Boosted Infrastructure

The transformation is not happening only in the visible ad layer, but also in the infrastructure that powers programmatic auctions across display, video, CTV, audio, and retail media. Generative models and more advanced machine learning systems are being used to:

• optimize traffic routing between SSPs and DSPs;
• forecast revenue and yield with more granularity;
• model audiences in environments with fewer direct signals;
• improve real-time bidding decisions.

On the other hand, the growing volume of synthetic content and sites created almost entirely by AI increases the risk of fraud, signal degradation, and low-quality inventory. This is pushing part of the market to focus more on transparency, cleaning up the media supply chain, and validating audience data.

For media buyers, the takeaway is straightforward: automation is getting more powerful, but it demands a stronger layer of governance, inventory quality verification, and caution around AI-generated environments.

Consumers, AI Content, and the Importance of the Human Touch

A recent Canva study on marketing and AI in 2026 reveals an interesting contrast. On one hand, users see value in personalization, relevance, and convenience when AI is used well. On the other, a large share of people say they can recognize AI-generated ads because they feel emotionally empty and lack authenticity.

Some of the points raised:

• about 70% of respondents say they notice when an ad seems machine-generated because it lacks soul, nuance, and originality;
• there is specific discomfort with social posts, product images, voice-overs, and emails that sound overly standardized;
• users are more open when AI is used to adapt messages to context, adjust language, improve accessibility, or simplify tasks, as long as it does not invade privacy.

Many people also argue that brands should be transparent when they use AI in communication and should reinforce data protections. The message is that automation does not replace creative vision, curation, and human sense of timing.

For brands and creative teams, this reinforces the idea that AI works best as a tool, not as a full replacement. Combining human direction, message depth, and technical structures tuned for algorithms tends to produce more sustainable campaigns.

AI Costs, ROI, and a Colder Corporate Look at Hype

While the hype is still strong, corporations are starting to look at the spreadsheet with a colder eye. Cases of companies signing up for every AI solution they could find, stacking premium models for basic tasks, are now hitting budgets hard.

Recent reports mention:

Tools we use daily

Translation

Text Inspection & Clipping

Social Media

Search & Research

Recruitment

Programming

Productivity & Organization

Operations

Marketing & Sales

Finance & Investment

Education

Design & Media

Data & Analytics

Content & Writing

Automation

• cuts in licenses for more expensive models, such as AI-based coding tools, due to a lack of clear return;
• executives questioning whether infrastructure and token spend is actually delivering productivity at the edge of the business;
• pressure for stricter governance, with concrete objectives for AI usage instead of diffuse adoption.

In this context, some model providers are adjusting their strategy. Anthropic, for example, released an updated version of its flagship model focused on balancing cost, speed, and reasoning depth, with faster, cheaper modes, workflows with multiple subagents, and control over computational effort. Chinese company DeepSeek, meanwhile, announced a permanent 75% price cut for its main model, backed by progress in domestic hardware.

For marketing teams, this opens room to test more tool options and fine-tune how much automation makes sense at each stage, always watching cost per outcome, not just chasing the latest tech buzz.

AGI Horizon, Regulation, and the Impact on the B2B Market

While the day-to-day keeps shifting, the long-term debate is also heating up. AI research leaders, like the CEO of Google DeepMind, argue that society has only a few years to prepare for systems with AGI-like characteristics, with 2029 showing up as a plausible date in some projections.

This discussion comes with warnings about:

• increasingly autonomous software agents as a dress rehearsal for more powerful systems;
• the ability of advanced models to improve their own development process in a self-accelerating loop;
• impacts on regulation, the job market, and public trust in technology.

In B2B, the trend is toward faster adoption of hybrid buying flows, where customers use agents to research, compare vendors, and negotiate conditions, while humans focus on strategic decisions. Vendors selling to businesses need to understand how these agents participate in the journey and how to make their offers easy for them to evaluate.

Social Networks, Subscriptions, and Prompt-Guided Personalization

Major social platforms are also moving forward with AI integration and new revenue models. Meta started offering subscriptions on Instagram, Facebook, and WhatsApp, including specific tiers focused on AI tools for creation, analytics, and personalization. There are layers designed for consumers, creators, and businesses, with benefits such as extra visibility, richer audience data, and advanced media features.

In the video space, YouTube began testing personalized feeds generated from simple text prompts. Instead of relying only on watch history, users can describe the type of content they want to see, and the platform assembles a continuous feed based on that explicit intent. This approach is very similar to how people use large language models, but applied to video curation.

For content creators, this changes the optimization angle: beyond pleasing the algorithm, it becomes useful to think about how content is described and discovered through natural language, which can influence titles, descriptions, and even script structure.

Brands, AI, and the Balance Between Technique and Trust

The through line connecting all these changes is relatively simple: the battle for attention and transactions is increasingly mediated by Artificial Intelligence, and this demands two capabilities from brands at the same time:

• technical excellence in SEO, structured data, metadata, media infrastructure, and integration with agents; and
• brand strength, with clear narratives, consistent experiences, multi-channel presence, and content that feels human, relevant, and trustworthy.

Brands that manage to combine these two fronts are more likely to stand out in an environment where it is not always the person who sees your offer first, but an AI system deciding what is worth showing.