August 7th-August 13th // Estimated Reading Time: 9 minutes
In This Edition!
Perplexity blocks Time from serving ads to its AI crawler
Giving people AI is cool, but giving them AI and training them on how to use it is even cooler
Seats Are Filling Up For The September AI Accelerator
The most popular AI bootcamp in the marketing industry is back! The AI Accelerator helps marketers level up on all things AI for marketing and advertising 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🔥
Perplexity blocks Time’s ads served to AI agents, calling them ‘deceptive’🔒
Source: Digiday
August 11th, 2026
Summary: Time set off an industry debate when it recently started running paid ads formatted as FAQs on the text-only pages read by AI crawlers. Perplexity says it's blocking those ads from influencing its AI knowledge base, warning that publishers using this format risk a downgrade in Perplexity’s trust score. Ally Bank and the Project Management Institute are testing Time’s ads, but many media buyers are split🔒 on whether the format would even work in improving their AI search visibility.
Opinion: This is another move in the ongoing game of cat and mouse between publishers and LLMs. They’re in direct competition with each other while also depending on one another. It’s an unhealthy relationship. Sure, LLMs want to make sure the information they surface to users is “unbiased” (whatever that means), which is their reasoning for shutting this down, but we suspect their bigger concern is around monetization. After all, how will an LLM ever go to an advertiser and say “pay me to show up in AI” if advertisers are already paying publishers to show up in AI?
Marketers are told to use AI, but no one's teaching them how🔒
Source: Adweek
August 11th, 2026
Summary: A new NewtonX survey of marketers found that 69% of companies mandate or push AI use, but only 43% of marketers received any formal training, and just 10% called that training adequate. The gap starts at the top: 81% of respondents said their companies overstate their AI proficiency. CMOs scored lowest🔒 of any level in the org. A separate Gartner poll backs this up: Only 15% of CEOs trust their CMO to lead on AI in 2026. NewtonX also found that 25% of marketers deploy AI tools without testing them first, and 4% publish AI-generated content without any human review.
Opinion: You can't half-ass AI transformation. That's what most companies are doing right now. You have to go top-down AND bottom-up.
Training teams on AI remains one of the biggest, untapped levers for AI transformation (no, we did not pay NewtonX to do this research). Don’t worry, we know a guy…
New Products & Features 🚀
What It Does: Google Ads got a home screen revamp with personalized AI insight cards and a prompt box for custom questions. Google Ads also launched Dashboards, which turn text prompts into visual reports with real-time explanations. Google added AI Overviews to the Analytics homepage that summarize performance changes since your last login. There's also a new benchmarking tool in Analytics that compares your campaigns against anonymized data from similar businesses.
Quick Take: Google adding helpful AI tools in its products makes a lot of sense. But marketers should be wary about how their interactions with these AI tools could be used. If you’re asking Google Analytics AI “I have $500k of extra budget, where should I put it?”, couldn’t that information get sent directly to your Google rep?
What It Does: Clinch and Yahoo DSP are partnering to let advertisers build one campaign, upload creative for two audiences, and let the platform pick the right ad for the right person in real time. Campaign setup also syncs automatically between the two platforms, so creative doesn't need to be manually trafficked.
Quick Take: This concept isn’t new. We’ve been doing the “right ad, right person, right time” thing since the dawn of digital. But AI makes it much more realistic, so we should absolutely rethink it, rewire it, and retest it.
What It Does: Anthropic is adding invisible watermarks to Claude generated text that tools can detect to comply with the EU AI Act. The watermarks will apply everywhere Claude is used: the API, Claude.ai, Claude Code, Claude Cowork, and Claude Tag.
Quick Take: AI disclosure is a work in progress. Watermarks are a good first step; they can survive light edits, but heavy rewriting, translation, or file conversion can strip them out.
AI Use Case of The Week💡
Mod Op built its own AI infrastructure, dodging a $3M licensing bill
The Setup: Mod Op is a 500-plus person independent marketing agency. Licensing off-the-shelf tools like Microsoft Copilot or Figma Weave for its entire staff would have cost an estimated $3M. Leadership also wanted to own its data and control the client experience, not hand both over to an external platform.
The AI Solution: Mod Op built Orion, an internal system connecting multiple LLM APIs to its own secured data. CTO Tessa Burg describes it as a combination of licensed tools and proprietary software running as a single platform. Internally, Orion's agents handle tasks like managing contractor workflows, prioritizing an employee's day based on their inbox, and building PowerPoints. A client-facing layer launched last year, giving clients a shared workspace for performance data, AI search visibility, and ad testing.
The Results:
• Mod Op avoided an estimated $3M licensing fees by building instead of buying seats.
• Client discovery, event resourcing, and strategy scoping dropped from 20-30 days to one week.
• Reporting and analytics work that used to take weeks now happens in real time.
Why This Matters: Mod Op's story could be helpful in better understanding the actual value behind an agency’s AI pitch. If you work with an agency claiming AI-driven speed or efficiency gains, it's worth asking what's driving that claim: licensed tools, custom workflows, or something built in-house. The latter could be significantly more differentiated and cost-effective than the former.
Your Action: When considering an agency AI pitch, don’t ask what the AI does, ask three things instead.
One: if you fired this agency tomorrow, could you easily reproduce the AI capability on your own?
Two: where does your data go, into infrastructure they control, or into some third-party vendor’s model? If it's the latter, there's no moat. You’re probably paying a markup for nothing.
Three: can we see the actual workflow? Not the deck, not the case study slide, but the steps, the integrations, the guardrails.
Proceed cautiously.
Other Notable Headlines📌
Gemini app hits 1B monthly users - Google's fastest-growing product ever now generates 150M+ images daily, and 63% of users talk to it instead of typing.
OpenAI is coming for SMB advertisers🔒 - The company is hiring for a dedicated SMB ads unit while also rolling out new measurement tools🔒 like conversion tracking and location-based targeting. Gaps in brand-safety verification and prompt-level data are still holding bigger advertisers back.
Why The Arena Group has, implausibly, rebranded into an AI company🔒 - The publisher holding company’s traffic fell 27% and revenue halved to $22M. Instead of a turnaround plan, it acquired an AI content generator and rebranded as Paradium.AI, drawing comparisons to Allbirds' and BuzzFeed's rocky AI pivots.
AT&T is betting big on open-weight AI models🔒 - The telecom giant uses 45B AI tokens daily and plans to shift 70%-80% of its AI usage to open models, which have already cut costs by 80-90% in some cases.
Target hires its first chief AI officer🔒 - Chandhu Nair joins from Lowe’s, where he oversaw data and AI innovation. Target sees AI as critical to its turnaround, including for improving sales forecasts and customer experience.
AI and organic search are doing different jobs, Shopify data shows - AI converts nearly twice as well as organic search when shoppers are comparing products, while organic search still wins when shoppers already know what they want.
Meta's cloud ambitions face cost and trust hurdles🔒 - Meta is considering renting out its AI compute capacity to offset $130-145B in 2026 capital expenditures. But analysts say Meta lacks the enterprise sales infrastructure and faces a data-privacy trust deficit compared to AWS, Azure, and Google Cloud.
OpenAI loses revenue chief Denise Dresser, second major executive departure in days - Dresser was CRO less than a year, and will be replaced by Dali Rajic, the former president and COO of cybersecurity firm Wiz. OpenAI COO Brad Lightcap is also leaving.
New York Post launches ‘Hamilton’ AI chatbot - Hamilton, named after the paper's founder, Alexander Hamilton, is an umbrella brand for AI tools like a chatbot, personalized briefings, and content recommendations, built on Google Cloud's Gemini platform and pulling from the Post's news archive.
Kochava's StationOne gains 5 ad platform workspaces for chat-run ad ops - Kochava rolled out chat-driven Workspaces for Google, Meta, Reddit, Snap, and TikTok, letting teams launch campaigns, run bulk edits, and pull reports through natural language instead of five separate platform UIs.
The future is for everyone - Mark Zuckerberg published a lengthy manifesto which says Meta will release more open-weight models and create a $1B fund for communities near data centers.
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!
