BRANDS
Pet food marketers on AI and marketing
Industry trade publication Pet Food Industry reviewed recent polling data which showed that marketers are moving slowly to adopt AI as part of their marketing product mix with over half of respondents still on the AI ‘sidelines.’
Nevertheless, there was clear adoption for some marketers as reporter Lisa Cleaver offered her three takeaways:
“1. Adoption remains limited but strategic. AI adoption in the pet food space remains in its early stages, with more than half of respondents still exploring or planning their AI marketing strategies rather than actively implementing them. This suggests companies are prioritizing risk management over rapid deployment.
2. AI helps content creation become more efficient. Pet food marketers who have implemented AI tools report content production acceleration as their most significant measurable impact, with 15% citing faster content creation as their primary benefit.
3. Data analysis trumps creative tasks. Among companies implementing AI marketing tools, customer insights and data analysis dominates usage at 21% of all respondents, suggesting those who have embraced the technology are focusing on data-driven applications rather than creative functions.”
Read more. (September 23)
Related: [Hotel industry] Guest segmentation in 2025: Moving beyond broad-brush marketing with AI (September 23) – hospitalitynet
BRANDS
Google and AI-enabled performance advertising
Google Ads’ Demand Gen campaigns (see more) are evolving, according to Search Engine Land. (September 23)
Google describes the Demand Gen campaigns product as a form of social advertising that captures “engagement and action across YouTube, including Shorts, Discover, Gmail, and Google Display Network.” It heavily leverages AI.
Search Engine Land outlined ‘What’s new’:
Target CPC bidding: Advertisers can now align Demand Gen with social campaigns for apples-to-apples budget comparisons.
Channel controls. Run ads only on YouTube, or expand to Display, Discover, Gmail, and even Maps.
Creative tools. Features like trimming and flipping help repurpose social assets for Shorts and other placements, lowering barriers to entry.
Feeds + app support. Product feeds in Merchant Center show a 33% conversion lift; Web-to-App Connect now extends to iOS for smoother in-app conversions.
Read more. (September 23)
More: Introducing Demand Gen Drops to help you stay on top of what’s new and drive results. (September 18) – Google’s Ads & Commerce blog
From tipsheet: Note Google’s tilt toward performance marketers with this product which echoes Meta’s focus in its advertising products such as Advantage +. AI loves the data-rich environment of performance marketing with clear outcomes in comparison to brand advertising.
LLMs & CHATBOTS
Developments
- Why OpenAI is building out a stacked marketing roster—and what its strategy could look like as it prepares to ramp up media buying and in-house creative capabilities. (September 23) – Ad Age (subscription)
- OpenAI Teams Up With Oracle and SoftBank to Build 5 New Stargate Data Centers (September 23) – Wired
- OpenAI CEO Sam Altman on creating “a factory that can produce a gigawatt of new AI infrastructure every week” with more details to come this year. (September 23) – Sam Altman blog
SELL-SIDE
Microsoft takes lead on publisher marketplace
Microsoft will imminently launch a pilot program called “Publisher Content Marketplace (PCM)” with a small group of publishers, according to Axios.
Unnamed publishing executives told Axios:
“- [Microsoft] eventually hopes to expand the pilot to more partners over time, working in tandem with them to build out the tools, policies and pricing models that could work for PCM.
– Microsoft’s Copilot assistant will serve as the first AI buyer within the marketplace for publishers to sell their content. The tech giant is looking to grow the marketplace’s demand-side offering to include other AI products.
– Microsoft discussed the pilot program last week at its invite-only Partner Summit in Monaco.
– One of Microsoft’s slides that caught attendees’ attention for its candor, and stood in contrast to other tech companies’ policies, read, ‘You deserve to be paid on the quality of your IP.’”
Read more. (September 23)
From tipsheet: This appears to be the next step of a plan that Microsoft started last year when it began to compensate publishers for AI usage. Flat fee, AI licensing deals for AI crawling privileges are numerous with usage-based deals still to come.
Other notable publisher AI efforts include Cloudflare announcing the private beta of its pay-per-crawl toll program in July. And the IAB Tech Lab has taken a leadership role with its Content Monetization Protocol (CoMP) effort.
AGENCIES
Agency: Podcast creators need to label AI content
Podcast ad agency Adopter Media says it is time for industry standards regarding content produced by AI. The company wants clear labeling and considers it a matter of “responsibility” for creators. See the company’s blog post. (September 23)
Industry trade publication InsideRadio covered the news and explained:
“The podcast industry has been buzzing with the AI debate this week after the work of Inception Point AI made headlines. It boasts that it can use AI to create episodes for one dollar or less, depending on how long and complex the episode is. The company’s Quiet Please Podcast Network has already produced more than 5,000 shows to date.:
Read more in InsideRadio. (September 23)
From tipsheet: There are implications here for all publishers who are considering AI-powered content and journalism. (I am not AI-powered, fwiw.)
AGENCIES
Ideas matter most, not AI
Barry Lowenthal, president of AI-enabled ad platform Inuvo, tapped his Media Kitchen agency roots in an Ad Age op-ed titled, “Why ideas, not AI, will decide who survives in 2030.” Read it. (September 19)
On LinkedIn, Mr. Lowenthal teased the premise:
“In a world where every ad element can be engineered for conversion, the only true differentiator left is ideas. AI will transform marketing services, but originality—zany, emotional, human ideas born from lived experience—will decide who thrives.”
Read a bit more on LI. (September 22)
Related: AI and marketing in 2030: actions to take now (September 23) – Ad Age (subscription)
TECH
Tracking outcomes online and off
The Trade Desk announced that it will use data management and intelligence platform Acxiom and its True Intelligence product to improve its ad platform’s measurement and track outcomes.
Acxiom is owned by agency holding company IPG.
AdExchanger’s James Hercher parsed the news yesterday in an interview with Acxiom global president Sean Muzzy and The Trade Desk CRO Jed Dederick:
“Since 2021, IPG has been a closed operator for TTD’s Unified ID 2.0, which is an entity that’s authorized to use first-party data to create and encrypt UID2 identifiers.
The new product goes further by plugging into The Trade Desk’s AI-based Kokai ad platform to create a live feedback loop between the UID2 IDs targeted by TTD and results collected by Acxiom.
It matches audience impressions to online and offline sales from Acxiom’s consumer data, Sean Muzzy, Acxiom’s global president, told AdExchanger.”
Read more. (September 23)
More: Acxiom Launches ‘True Intelligence’ to Revolutionize Digital and CTV Advertising Measurement in Partnership with The Trade Desk (September 23) – press release
Define: agentic advertising
Yesterday, Scope3 CEO Brian O’Kelley continued his company’s campaign to support grassroots standardization efforts for agentic advertising.
Urging feedback on his latest LinkedIn post, Mr. O’Kelley defined ‘agentic advertising’ and then threw in a couple of other definitions for extra credit:
“- Agentic Advertising is enabling agents and their human counterparts to execute ad campaigns
– Agentic Media is running ads on AI chat environments
– AI Syndication is selling content to AI companies for indirect monetization”
Feedback in Mr. O’Kelley’s LinkedIn post comments included…
- Amazon’s global head of ad tech services Eran Metzer: “I like the approach, but I think it extends beyond just executing ad-campaigns. I run into agentic workflows in advertising that focus on tasks such as budget, planning, measurement, verification and others that impact advertising.”
- Marketecture founder, and former AppNexus executive, Ari Paparo: “I like your definition of Agentic Advertising. One hopes this is not just automation of existing ad tech workflow but actual intelligence on behalf of brands. I don’t love the phrase “Agentic media”. If I think of an ad within a chat bot response that seems like something simpler like an “AI Ad” or to borrow from search we can call it AI Marketing or AIM.”
- Former Google and MediaMath executive Frederic Lutt said: “For me, Agentic Media would define the advertising running on AI agents using the AI models (like a network of 3rd parties). If the advertising is displayed directly on ChatGPT, Gemini etc…, then Conversational Media might be a good option. The user is having a conversation when the ad is displayed. As a French person, I like Agentic Media as it is easy to translate in French: Media Agentique. Conversational Media would be “Media de Conversation” and it is more difficult to pronounce!”
- Head of EU Amazon ad tech solutions, Tyler Swarz said: “A fourth option we as an industry need to consider is ‘marketing to machines.’ As agentic browsers do more shopping on our behalf, is there any way to influence what is surfaced to the end user?”
See over 40 comments and counting on LinkedIn. (September 23)
From tipsheet: As one might expect, there has been no input from the independent large language model (LLM) companies such as OpenAI and Anthropic. They don’t have ads embedded in their chatbots yet. In time, along with the ‘walled gardens’, their ‘takes’ on agentic advertising might be illuminating considering their AI-native experience.
TECH
AI for CTV: Optimization over creative
Noting the “hype” around AI and ad creative and related announcements, tvScientific co-founder and CEO Jason Fairchild agrees that making CTV creative more accessible to SMBs and performance marketers is a step forward.
Yet, Mr. Fairchild believes there’s a more important key to success in CTV — which also speaks to tvScientific’s strategy. He explains, “Creating a TV commercial with AI is very different from using AI to make that ad spend perform. That’s the actual challenge and the biggest opportunity.”
Read Jason Fairchild’s Inside Performance Advertising newsletter. (September 23)
HR
Now hiring
AI-enabled media planning platform QuantumPath announced former AudienceScience CEO Jeff Hirsch as its new Chief Executive Officer (September 23) – press release
PROMPT
Explain the specifics of how AI agents used for advertising will likely change in 5 years.
Response from Mistral’s Le Chat:
- AI Agents as Consumer Proxies
- AI-to-AI Persuasion and Negotiation
- Hyper-Personalization and Contextual Relevance
- Autonomous Campaign Management
- Trust and Transparency as Competitive Advantages
- Real-Time, Dynamic Ad Experiences
- Shift in Media Planning and Buying
MORE
- Google Defends Ad-Tech Business, DOJ Witness Urges Breakup (September 23) – MediaPost
- Rafael Nadal complains of AI advertising scam using his image (September 23) – The Economic Times
- How AI decisioning will change your marketing (September 23) – MarTech
- Appier Announces Full Product Line Infused with Agentic AI, Ushering in a New Era of ROI Driven Marketing Solutions (September 23) – press release
- There isn’t an AI bubble—there are three (September 22) – Fast Company