August 14th-August 20th // Estimated Reading Time: 8 minutes
In This Edition!
Who should do AI transformation for marketers, consultancies or agencies?
GEO for Alexa!
Seats Are Filling Up For The September AI Accelerator
The AI Accelerator (sponsored by Blis) helps marketers level up on all things AI for marketing and advertising, like planning, creative, activation, and measurement, so they can make a meaningful impact on their business, fast.
We only have 40% capacity left! Enroll now using promo code EarlyBirdsLearnAI for 20% off!
Top Headlines🔥
PepsiCo is conducting an AI marketing transformation review🔒
Source: Ad Age
August 17th, 2026
Summary: PepsiCo wants to overhaul its marketing operations to run leaner and smarter with AI. Omnicom, Accenture, Deloitte, and Publicis's Sapient unit are among those pitching for the business. Omnicom is PepsiCo's agency of record.
PepsiCo CEO Ramon Laguarta told investors earlier this year that half a decade of work upgrading the company's data and cloud systems has set the stage for this next phase. PepsiCo is already using AI in marketing to sharpen consumer insights, generate content, and personalize outreach.
Opinion: On paper, consultancies are better positioned to win these kinds of “transformation” pitches. They know how to restructure organizations, rebuild workflows, and bill for this work; that's their whole business model. But agencies might be better suited to do the job, because it’s all about marketing transformation.
Marketing is a unique beast. There’s so much that goes into it: data management, audience research, segmentation, media planning, activation across dozens of platforms, creative production, measurement, and much more. Each discipline has its own tools, data flows, and constantly shifting externalities (e.g. signal loss, walled gardens, platform algorithm updates). Knowing where AI affects the tooling itself is a critical aspect of the transformation work, and it can guide the workflow and human transformation. That's Omnicom's and Publicis's edge: decades of running actual campaigns through this mess and being on the forefront of how AI is changing everything in marketing.
Amazon's AI shopping assistant is narrowing the shelf brands compete for🔒
Source: Ad Age
August 17th, 2026
Summary: Amazon's AI shopping assistant, Alexa for Shopping (formerly known as “Rufus”), now gives shoppers just one or two product recommendations instead of a full page of results. There’s also personalization: different consumers may receive different recommendations. That has brands scrambling to figure out how to be the one Alexa recommends. Brands are rewriting product data, brand store content, and even their DTC sites and PR content to boost their odds of getting recommended, essentially mimicking ChatGPT-style AEO and GEO tactics. Amazon has also cut product titles down to 75 characters (from 250 characters), pushing brands to move detailed information into other parts of the product page instead.
Opinion: This seems like a huge shift from traditional Amazon search, which used to show you page after page of products. But the goal of old-school Amazon search was the same — be at the top of page one. Whatever gets you there probably gets Alexa to recommend you too. Our conclusion: on Google, GEO (AI visibility optimization) as a practice can be very different from traditional SEO (search engine optimization), but on retail sites (like Amazon) perhaps it’s more similar.
New Products & Features 🚀
What It Does: Meta AI's new Mac app, powered by its Muse Spark model, lets users talk to their desktop apps and read the screen to answer questions. It ships alongside a broader small business update, letting merchants connect Instagram, Facebook, ad accounts, and Google Workspace for performance insights, competitor intelligence, and auto-generated decks.
Quick Take: Meta CEO Mark Zuckerberg has said selling AI agents to businesses is a major opportunity. This update is one on-ramp: Give small business owners a reason to open Meta AI regularly, then convert them to the paid tier once the free version becomes part of their workflow.
What It Does: Similarweb's new AI Ads dataset shows advertisers what ads are running inside ChatGPT, Google AI Mode, and AI Overviews, and how they stack up against competitors. The data is pulled from real user panel conversations. Similarweb says 26% of ChatGPT responses now carry an ad, and nearly 30% of AI Mode responses now carry an ad.
Quick Take: Understanding your competitors’ AI ads presence helps you run smarter AI ads. The industry’s infrastructure for running AI ads continue to grow every day. It’s obviously going to be the next big ad channel.
AI Use Case of The Week💡
Shoppers buy more after Albertsons consolidates its AI shopping tools🔒
The Setup: Albertsons expects sales to fall this year after its core grocery business struggled last quarter, as shoppers turned more cautious. The company is leaning on better AI-powered shopping experiences to win those customers back.
The AI Solution: Albertsons previously introduced Ask AI, a conversational agent in the search bar of its websites and later added Plan AI and Buy AI to handle other parts of the shopping journey. It has consolidated all three into a single assistant that provides a more comprehensive experience to help shoppers find products, surface related items, and build shopping lists for occasions like a dinner party.
The Results:
• The consolidated assistant drives a 26% lift in average order value, compared to 10% lift from Ask AI (just conversation search)
• Consolidating the tools means separate teams are no longer spending time and resources on duplicative tools
Why It Matters: AI tool sprawl is real. And shoppers are starting to expect planning capabilities, discovery, comparison, and guidance from AI, not just simple search assistance. A customer planning taco night shouldn't just get pointed to tortillas; they also need to get walked to the cheese, the vegetables, and everything else they'd normally forget. And they shouldn’t have to use multiple AI tools to get there.
Your Action: Do your customers interact with multiple AI-powered tools across your site and app? If the answer is yes, you likely have Albertsons' problem: customers switching between disconnected assistants instead of having one continuous conversation. That hurts customer experience and ultimately hurts revenue. Walk through your own customer journey and note every point where you have to leave one AI tool and start over in another, from search to recommendations and checkout. Each handoff is a chance for customers to drop off or get frustrated. Prioritize merging the two tools with the most overlapping traffic first and measure basket size or conversion before and after. That's the fastest way to prove value.
Other Notable Headlines📌
SpaceX officially closes its $60B Cursor acquisition - Cursor, the AI coding platform, now officially has the same parent company as X: SpaceXAI. And the same boss…
Google buys Spirit Airlines' data for $10M to train its AI models - The deal, part of Spirit's bankruptcy sale, gives Google access to internal emails, HR records, booking and frequent flyer data, but no personally identifiable info. A bankruptcy judge will rule on whether the deal can go through this week.
OpenAI builds ChatGPT ads in Europe on consent, not legitimate interest🔒 - Users who opt out of personalized ads still see contextual ones, enabling ChatGPT to avoid the $451.9M GDPR fine that Meta had to pay.
IAB revisits AI disclosure in ads as legal requirements multiply - The revised framework helps marketers adhere to new AI labeling laws in New York, California, the EU, and Asia, using tools like a standardized sparkle icon or text label.
Meta brings Pocket, an app that lets you vibe-code and share games, to US users - The app lets people use AI prompts to generate small, interactive games and post them to a shared feed.
Omnicom offloads hundreds of staffers who built its AI platform to a third-party contractor🔒 - Omnicom is saying publicly that its AI platform is central to its future, while quietly gutting some of the internal team that built it.
'Let's pee on computers': New beer ad taps AI data center backlash - The joke commercial from Liquid Death and Garage Beer, featuring Jason Kelce, proposes using human urine to cool AI data centers amid growing public concern over their water use.
That’s It For This Week 👋
The U of Digital AI Edge Newsletter is intended for AI Edge Pro subscribers, but occasional forwarding is okay!
To subscribe visit Uof.Digital/Newsletters/ai-edge or AI Edge Pro, or contact us directly for group subscriptions.
And remember, U of Digital helps teams drive better outcomes through structured education on critical topics in marketing / advertising like AI, mar tech, programmatic, CTV, commerce media, and more. Interested in learning more about how we can supercharge your team? Let’s talk!
Thanks for reading!
