On today’s EMARKETER Miniseries—AI-Driven Media Management—we explore how to break down the media manager role into workflows that can be automated or augmented by agentic AI, what agencies misunderstand about AI, and which agency tasks are ripest to hand off to AI right now. EMARKETER Senior Director of Content Jeremy Goldman speaks with Adam Epstein, co-founder and CEO of Gigi. Listen everywhere you find podcasts, and watch on YouTube and Spotify.
As retailers prepare for next year, they acknowledge that convenience has evolved from a value proposition to a structural shift in how all of retail operates. We asked leaders across retail media, digital identity, payments, mobility, and connected commerce, and they agreed that convenience will continue to change throughout the next year as expectations shift and AI eliminates friction.
2025 challenged many of retail’s long-held assumptions. What looked like familiar patterns often turned out to be something different entirely, and in the process, a few key trends were either missed or misread by brands trying to make sense of shifting shopper behavior. Here are three trends from 2025 that were either overlooked or misunderstood, and why they will matter in the year ahead.
Google is testing a new AI agent, CC, to change how the workday begins. Instead of another app or dashboard, CC delivers a daily email briefing that summarizes information pulled from Gmail, Calendar, and Drive. As AI agents begin ranking and summarizing messages, the inbox shrinks and attention gets scarcer. Marketers who rely on email will need to understand tools like CC and optimize for agents—not inboxes—as AI assistants decide what’s worth surfacing. Catchy subject lines will no longer be enough to earn attention.
Integral Ad Science is expanding far beyond verification with IAS Agent, a DSP-embedded AI assistant that analyzes live campaign data and autonomously optimizes safety, suitability, and contextual settings. Built on IAS’s multimodal dataset, the tool evaluates trillions of signals across text, video, audio, and imagery to refine targeting decisions faster than human workflows—early testers reported efficiency gains and reduced waste. IAS frames the product as a step toward agentic advertising, enabling real-time optimization, natural-language insights, and interoperability with other AI systems. The launch comes as private equity bets heavily on independent measurement players amid rising demand for precision and transparency.
Over one-third of Gen Z (39%) and millennials (34%) who have used genAI tools to check symptoms report that they would put off seeing a doctor if the AI told them their issue was low-risk, according to an October 2025 poll from The Mesothelioma Center at Asbestos.com conducted by SurveyMonkey. Overrelying on AI for medical guidance carries real risk, especially as models are still maturing and sometimes produce faulty information. AI companies should add explicit in-chat disclaimers against being used as a replacement for medical care and strengthen guardrails to block unvetted or potentially harmful health advice.
DoorDash is testing an AI-powered app that helps consumers find new restaurants to explore, expanding its business model beyond food delivery. Zesty pulls in signals from across the web and provides citations like Yelp reviews, Google Maps ratings, and mentions on Reddit threads, per Bloomberg. Brands should focus on both generative engine optimization and earned, shared, and owned (ESO) media placements to maximize the chances of showing up in AI responses. Monitor brand mentions and reviews on ESO channels to address any negative responses from customers and ensure accuracy of business details.
Excitement about AI and its enterprise possibilities aren’t translating into meaningful adoption as corporate strategies lag. More than one-third (38%) of tech leaders say their organizations are piloting agentic AI projects, but only 11% have agents in production, per Deloitte’s Tech Trends 2026 report. Foundational challenges—such as data quality and security governance—are blocking the path to clear ROI. CMOs should balance excitement about agentic AI’s potential by first applying the technology to larger pain points. Establish clear KPIs before deployment, and monitor the quality of initiatives to ensure that initial goals are met.
Amazon is in early talks to invest up to $10 billion in OpenAI, three sources told The Information. If finalized, the deal would value OpenAI at more than $500 billion and intensify Amazon’s competition with Microsoft in AI cloud services. The investment, which follows November’s $38 billion deal between the companies, would bring OpenAI deeper into Amazon’s ecosystem to use Amazon’s Trainium chips to develop future AI products. Consolidation means faster model updates, tighter cloud integrations, and more reliable access to genAI tools. Choosing the right AI tools and cloud providers now becomes a turnkey solution.
"The retail media landscape is only becoming more crowded, but Target's guest insights are often cited as a key differentiator," said our analyst Sarah Marzano during EMARKETER's recent Commerce Media Summit.
As consumers grow more comfortable with using AI, retail industry leaders see 2026 as a pivotal year in shaping how the emerging technology disrupts the way people shop.
Later has transformed from a scheduling tool into a full-scale creator-commerce engine. One year after acquiring Mavely, the combined platform is processing more than $2.4 billion in annualized GMV and has paid out over $250 million to creators. During Black Friday–Cyber Monday alone, creators drove $50 million in sales through Later and Mavely systems. With link-in-bio tools, affiliate rails, workflow software, and AI-powered attribution stitched into one stack, Later now acts as a performance channel for brands like Southwest and Bissell. Its EdgeAI engine ties creator posts to SKU-level results, reflecting a broader shift toward creator marketing as a full-funnel, revenue-driving discipline.
WPP Media and YouTube are expanding their partnership to bring non-public YouTube video and creator data into WPP’s AI system WPP Open, per a press release. Taking advantage of WPP Media’s offering gives advertisers the ability to partner with creators on the most popular social platform in a far more measurable, practical, and effective way than before.
Prominent clinicians and healthcare experts report a growing trend of bad actors using AI to impersonate them online and push unsafe products or unreliable medical information, according to a recent New York Times article. AI deepfakes may further discourage doctors from having their images and voices online. Social platforms must reassure healthcare creators about how they detect AI-driven scammers, enforce impersonation policies, and respond swiftly to deepfake reports.
Agencies are increasingly acting as commerce media guides, helping brands move past outdated structures, sort through measurement standards, and bring AI into their planning. While commerce media networks (CMNs) have expanded to capture more than just retail media dollars, the silos between brand, retail, and sales teams make integration a challenge.
Healthcare AI startup OpenEvidence is aiming to raise $250 million in equity funding, amping its valuation to $12 billion, which doubled from just two months ago, per The Information. OpenEvidence’s surge in valuation underscores physicians’ growing demand for AI tools that surface credible, peer-reviewed information.
OpenAI’s GPT-5.2 is its answer to advancing models from Gemini and what the company calls its best model yet for professional use. The latest model improves spreadsheet creation, presentation building, code generation, understanding of images and long contexts, tool use, and handling of complex projects, per an OpenAI blog post. As brands expand past using AI for image generation and into higher-level tasks, experiment with GPT-5.2 by testing each mode to see which offers the most dependable outputs for analytics, planning, and creative development.
AI is permeating retail and playing a substantial role in the purchase journey for many holiday shoppers this year, though many didn’t notice its presence. More than three-quarters (78%) of US adults said AI touched their holiday shopping experiences—including via “delivery notifications, return systems, or virtual assistants”—per Liveops’ Holiday AI and Customer Service report. Gen Zers led the charge with 89% using AI in some way during holiday shopping. Gen Z’s comfort with automation makes them a great testing group for AI copilots, and retailers should experiment with genAI-driven personalization, on-site product suggestions, and conversational assistance.
Meta is shifting strategy from mining its online social networks to tapping more real-world conversations as original posts on Facebook and Instagram fade, per Social Media Today. Its acquisition of Limitless and partnership with ElevenLabs open up new data channels for its AI models. Meta’s alternative AI training sources could deepen personalization and predictive insights if these conversation flows feed next-gen recommendation engines. Advertisers might gain richer behavior signals and context that go beyond scroll patterns and likes.
AI chatbot use has crossed from novelty to habit for US teenagers. Nearly half (46%) of teens ages 13 to 17 use chatbots at least once a week, per Pew Research, including 16% who access AI several times a day or almost constantly. A new generation of AI-native users is emerging and could soon define which chatbots and AI services become the industry standard. Marketers should pressure-test AI integrations for age-appropriate use cases. As teen AI use grows, safety-first design can earn trust and long-term loyalty.