Amazon is introducing “Help Me Decide,” an AI-powered shopping feature that recommends the best product for users comparing similar items. The tool analyzes browsing patterns, search history, and purchase behavior to offer personalized suggestions, along with explanations based on customer reviews and key features. Initially rolling out to millions of US shoppers, it joins Amazon’s growing lineup of AI-driven tools like Interests, Shopping Guides, and Rufus. While Amazon and rivals such as Google, Meta, and OpenAI are all racing to integrate AI into ecommerce, the industry is still testing which approach will truly enhance the shopping experience.
Alphabet subsidiary Verily is launching a free health app offering personalized guidance from clinicians. The Verily Me app will also have an AI agent to answer people’s health questions based on their medical records. Verily’s competitive advantage over bigger companies with brand-name is that it has clinician partners and access to some medical record data. The company should leverage its network of doctors to endorse Verily Me to their patients, using real-world examples to demonstrate the benefit of combining a person’s health history with a medical expert’s view for individualized guidance.
Reddit is suing Perplexity and data-scraping companies Oxylabs UAB, AWMProxy, and SerpApi, highlighting the battle over user-generated content (UGC) in the race to build the top genAI models. Cases like this could redefine how AI firms access and value online content, including original UGC and brand-owned material. To navigate an increasingly complex landscape for information sourcing, marketers should diversify reliance on genAI tools and explore AI partners that offer legal indemnification clauses to ensure that a brand isn’t at legal risk if a provider errs by scraping copyrighted information.
Privacy regulations and platform changes are creating blind spots for mobile marketers and forcing them to rethink how they attribute app installs and measure performance. Forty-one percent of mobile growth, marketing, and product leaders worldwide say privacy measures online are leading to difficulties with cross-channel attribution, per Branch’s 2025 State of App Growth report. Brands should unify marketing and product data into a single, trusted performance dashboard, make cross-channel measurement a strategic KPI, and use AI for contextual analytics and smarter targeting to find signal gaps left by privacy-driven data loss.
Artificial intelligence is transforming how brands navigate media buying, with digital ad buyers using AI for processes like ad personalization, audience insights, and creative ideation. In a conversation with EMARKETER, Mike Hauptman, CEO of cross-DSP manager AdLib, discussed how AI is altering the media buying landscape. Marketers are operating in a landscape where AI is a necessity—but as challenges are expected to persist for years to come, those who thrive will be the ones who find a happy medium.
Netflix is going “all in” on AI, making the tool a core part of how content experiences are built, customers are acquired, and ad campaigns are targeted and planned. It's focusing its AI investments on product experiences, content production, and advertising. Despite Netflix’s “all in” attitude, consumers are still hesitant about genAI content in creative fields, including genAI advertisements. CMOs should innovate responsibly by testing AI-driven creative tools where they can enhance performance while maintaining transparency and human oversight. Piloting AI in infrastructure tools—such as website search and customer service agents—rather than creative content could also help customers be more comfortable with its applications.
Meta will cut 600 roles from its Superintelligence Labs (MSL) division as it tries to move faster in the AI race. The layoffs are concentrated on its Fundamental AI Research (FAIR) unit, per Axios. Hiring will continue for Meta’s newly formed TBD Lab group, which has been poaching AI-skilled employees from OpenAI and Apple with splashy and costly buyouts. The rush to commercialize AI raises the need to invest in responsible research. Sustainable AI performance could depend as much on guardrails as it does on growth: CMOs should invest in AI tools that deliver measurable impact today while prioritizing vendors that vet tools, establish guardrails, and demonstrate accountability to protect both brand equity and consumer trust.
36% of marketers say user-generated content (UGC) is extremely important to their social media strategy, compared with 2% who say the same for AI-generated content, according to an August 2025 survey from PhotoShelter.
OpenAI's new Instant Checkout feature for ChatGPT allows users to purchase products directly through the platform without ever leaving the interface, potentially creating a new retail channel that could reshape online shopping behaviors. However, analysts remain cautious about its immediate impact.
Leading healthcare AI startups, including OpenEvidence, Abridge, UpToDate, and Doximity, are rolling out new products and capabilities in the race to compete for physician adoption and investment funding. Companies could gain an advantage by making their products easily integrated into clinicians’ existing workflows, such as their EHRs. Startups should also showcase the outcomes of their technology to influential stakeholders like medical associations to help establish credibility at the clinician level.
OpenAI launched an AI-powered browser—ChatGPT Atlas—and jumped headfirst into a new kind of rivalry with Google and Perplexity. ChatGPT Atlas is built around OpenAI’s flagship chatbot and features agentic capabilities. The browser is available globally on macOS, and access for Windows, iOS, and Android users is coming soon, per OpenAI. Companies should start optimizing for conversational search by ensuring websites are structured so AI agents can easily find and surface them in user queries. Brands should test both Atlas and Comet to see how their content surfaces and understand how AI browsers engage with users.
Airbnb opted against launching a third-party app integration with ChatGPT because it “didn’t think [the technology] was quite ready,” CEO Brian Chesky told Bloomberg. While the company hasn’t ruled out joining the platform, its measured approach contrasts sharply with the enthusiasm of competitors like Expedia and Booking.com. Airbnb is prudent to have reservations about OpenAI’s commerce capabilities. While chatbots could reshape travel discovery and booking, early reports indicate ChatGPT’s utility for now is hampered by a clunky, finicky interface that is more frustrating than helpful. Rather than forge commerce partnerships with AI companies, Airbnb is focused on making its service an indispensable travel resource to keep its platform sticky.
Now that consumers can make direct purchases within ChatGPT, marketers and retailers must reimagine the customer journey once again.
AdsGency, which bills itself as the first agentic operating system for advertisers, is working to unify the entire advertising process in a single ecosystem. Its large language models (LLMs) target the ideal audience, create the ads, and automate the ad-buying process. AdsGency is breaking down siloes and democratizing advertising for smaller teams that don’t have the talent budget of larger companies. But at the same time, it’s taking over for humans and can easily miss the nuance that people can provide. Brands could adopt systems like AdsGency for targeting, placements, and analytics but leave the content to human creatives.
Marketing measurement is entering a new phase of speed and precision. InMarket’s Michael Della Penna told EMARKETER that marketers are moving beyond static reports toward real-time insights—fusing marketing mix modeling (MMM) and multi-touch attribution (MTA) to understand what drives incremental sales as campaigns run. AI-powered models now forecast lift, optimize spend, and connect awareness to conversion through unified platforms. With 56% of marketers prioritizing sales lift and nearly half investing in MMM, the focus is clear: decision speed over dashboards. InMarket’s end-to-end system exemplifies this shift, reframing measurement as a continuous feedback loop rather than a quarterly report.
B2B buyers are leaning on AI tools for vendor selection, raising the stakes for surfacing in AI results. Eight in 10 global B2B buyers in the tech industry use genAI as much as traditional search when researching vendors, per Responsive’s Inside the Buyer’s Mind report. Four in 10 use genAI and traditional search equally. B2B marketers can insert themselves early in buyers’ discovery, vetting, and selection process by focusing generative engine optimization (GEO) efforts on controllable platforms. Ensure website information is structured and easy to parse by publishing clear FAQ pages with information on pricing, use cases, and product offerings.
Adobe’s new genAI model marketplace—Adobe AI Foundry—lets brands create bespoke versions of its Firefly AI model. The marketplace helps enterprise users train and deploy customized content-creation models by retraining Firefly’s base knowledge. The models can understand brands’ tone, style, and products, per VentureBeat, then generate content accordingly. Platforms like Adobe AI Foundry could help marketers create more relevant, personalized ad experiences across platforms. CMOs should treat model customization as a way to consolidate creative tools and vendors while scaling personalization at speed, using it to strengthen collaboration between in-house teams and agencies.
On today’s podcast episode, we discuss how digital has changed in 2025: why the digital ad triopoly (Google, Meta, and Amazon) are losing influence, how YouTube is still under valued, how AI search behavior is changing, and more. Join Senior Director of Podcasts and host, Marcus Johnson, VP of Global Content Operations, Eleni Digalaki, and Principal Analyst, Yory Wurmser. Listen everywhere and watch on YouTube and Spotify.
The vast majority—86%—of citations from AI-generated responses come from sources that brands either directly control or strongly influence, according to an analysis by Yext. That includes their own websites, listings, and reviews. It’s imperative for companies to act quickly to optimize their websites, product listings, and other content for AI-driven discovery.
Pinterest is giving users control over the flood of genAI content on the platform with a new tuner that allows users to control how much genAI content they see in specific categories, per a Thursday announcement. By giving users control over how much genAI content they see, Pinterest is creating a safer environment for advertisers, reducing the risk for brands by ensuring ads don’t appear alongside content that audiences dislike or want to avoid.