Amazon Advertising rolled out a new basket of AI-enabled tools across creative, media and planning at its unboxed event in Nashville, Tennessee yesterday.
Amazon Ads VP of Product Jay Richman outlined each tool on Linkedin yesterday:
“Campaign Manager: We’re bringing the Ads Console and Amazon DSP together into one unified media buying tool. Advertisers can now plan, buy, and measure campaigns through a single entry point, making ad management faster and more intuitive.
Ads Agent: An AI-powered assistant that supports advertisers at every step, helping them move faster, act with precision, and focus on what really drives results.
Creative Agent: This one’s especially close to my heart. It’s a generative AI tool that helps brands produce high-quality creative — including Streaming TV ads — that does product research, storyboard creation, and ad production, at no charge to advertisers. Designed to supercharge human creativity, it’s already producing some amazing results for customers.
Full-Funnel Campaigns: Our first AI-powered campaign type that recommends the optimal setup across Sponsored Products, Sponsored Brands, Display, and Streaming TV, making full-funnel advertising easier to launch and manage.”
Read more on LinkedIn. (November 11)
More:
- Amazon Ads showcases the future of advertising at unBoxed (November 11) – Amazon Ads
- “Amazon is launching AI agents into its DSP, it’s partly a move to simplify the DSP, which has sometimes been characterized as complex compared to rivals.” (November 11) – Ad Age’s Garett Sloane on LinkedIn
In related news, Amazon said that Prime Video now had the reach of 315 million uniques – up from about 200 million just over a year ago.
LLMS & CHATBOTS
Anthropic joins Amazon on stage
As part of the roll out of Amazon Ads’ new Creative Studio AI tools, Amazon VP of Producct Jay Richman will join Anthropic’s partnerships executive, Daniel Rosenthal on stage at unboxed today.
Livestream begins at 10 a.m. CT.
Is Amazon Ads getting tighter with the maker of the Large Language Model (LLM) behind the Claude chatbot?
Amazon is a strategic investor in Anthropic.
Read a preview of the creative tools in Fast Company. (November 11 – subscription)
LLMS & CHATBOTS
Developments
- Google is introducing its own version of Apple’s private AI cloud compute (November 11) – The Verge
- The Little-Known Search and Browser Startup (Brave) Generating $100 Million in Annualized Revenue (November 11) – The Information (subscription)
- Politics comes for American AI (November 11) – Politico
LLMS & CHATBOTS
AI company profitability plans
AI company Anthropic is on track to turn a profit in 2028, with competitor OpenAI projected to do so in 2030, according to documents seen by The Wall Street Journal.
The WSJ’s Berber Jin reported on Monday:
“OpenAI’s large upfront investment, particularly for new chips and data centers, could pay off handsomely if demand for its products continues to surge. The company recently launched a new video app called Sora and a web browser named Atlas. It is working on a new consumer hardware device, e-commerce and advertising features for ChatGPT, and humanoid robots.”
Read more in the WSJ. (November 10)
From tipsheet: The “advertising” revelation in the article seemed like a throw away line by the reporter with no reference as to how it was sourced. It’s not a foregone conclusion that ads are a part of OpenAI’s plans given almost all public comments by CEO Sam Altman to date.
BRANDS
Nielsen on AI coming to TV
On Mike Shields’ Next in Media podcast, Nielsen CEO Karthik Rao provided a look at his company’s evolution in the age of AI.
Speaking specifically about AI-enabled advertising, Rao saw opportunity ahead in TV. From the podcast:
MIKE SHIELDS: “AI buying has really taken off [with] platforms like Google and Pinterest… It hasn’t really happened in TV yet, but you gotta wonder if it’s coming. Do you think about that and what role Nielsen might play there?”
KARTHIK RAO: “Yes, I think it’s really important because AI is the one thing that is your best shot at bringing disparate worlds together. This whole excitement about AI, what changed? Because machine learning existed, deep learning existed for decades, and we used all those things -such as content recognition. So what changed?
Number one, the ability for models to take structured and unstructured data, which is the ground breaking thing. (…) And then the ability for it to self-learn as you keep pumping more data into it.
So that makes me particularly optimistic because it tells you that if you can bring a structured and unstructured world together, then you can also think of how that applies to many other contexts – like linear TV buying versus digital TV buying.”
MIKE SHIELDS: “All these things that are hard to mesh right now can theoretically be more meshable.”
KARTHIK RAO: “100%…”
Read more on Next in Media. (November 11)
And, hear the podcast on the Apple Podcasts app.
PLATFORMS
Innovid wants to be agentic connector
Mediaocean’s Innovid announced the launch of new AI agents alongside the company’s Innovid Orchestrator, which it describes as “the industry’s first full-cycle AI orchestration superagent built to connect people, data, and technology across the advertising lifecycle.”
Read the release. (November 11)
AdExchanger’s Victoria McNally covered the news and said that the agents “help automate basic advertising activities, including ad creation, delivery, measurement and optimization.”
Bayer’s head of investment for ad tech, Gary Guarnaccia, told McNally that the new product line is welcome: “The last thing marketers need is more complexity or closed systems.”
Read more in AdExchanger. (November 11)
In an Innovid blog post, president Grant Parker wrote:
“Think of Orchestrator as the ‘superhighway’ that connects it all. It gives every agent a common road to drive on. That means data, decisions, and outcomes can move freely and intelligently across the entire advertising lifecycle.”
From tipsheet: Orchestration makes sense given the breadth of Mediaocean’s offerings which pulse with advertising data. Prisma President Ramsey McGrory explained in July that Mediaocean comprises the traditional “back office” business now known as “Prisma” (was MediaBank/Donovan) as well as two “parallel” businesses: Innovid for transacting and video ad serving/DSP products and Protected for ad verification.
AGENCIES
IPG touts AI tools in earnings report
In advance of its merger with Omnicom, ad agency holding company Interpublic Group reported Q3 2025 revenues and profits that beat analyst estimates.
Read the report on the SEC’s website. (November 10)
The ad holding company said that the better-than-expected results were buoyed by spending from media and health care clients. Reuters added:
“Surging use of AI tools that let businesses create ads more cheaply and quickly has squeezed traditional agencies, prompting them to build similar in-house tools to retain clients.
As a result, IPG has integrated AI and data platforms such as ‘Interact,’ an AI-powered system that unifies data, creative and media for personalized marketing, into its workflows.”
Read more. (November 11)
More: IPG Q3 earnings show decrease in revenue, but increase in profit ahead of Omnicom acquisition (November 10) – Campaign US
From tipsheet: The finalization of IPG’s merger with ad holding company Omnicom is expected by November 28.
SELL-SIDE
New Wikipedia content marketplace for AI
Wikimedia Foundation – the steward for Wikipedia – rolled out a new platform for AI developers to buy and license content from the popular online encyclopedia. In a blog post on the Wikimedia blog, the Foundation pitched, “In a world increasingly awash with AI, Wikipedia’s human knowledge is more valuable to the world than ever before.” See the post. (November 10)
See: Wikimedia Enterprise platform. And, the “price list” for spidering content.
TechCrunch noted that Wikipedia’s human traffic is down 8% year-over-year as AI search has apparently had its effect. Read more.
PLATFORMS
Meta discusses GEM ad model
Meta revealed more details about its Generative Ads Model (GEM) which hones personalization and was publicly launched back in March.
The company claimed GEM uses the “largest foundation model for recommendation systems (RecSys) in the industry, trained at the scale of large language models.” Promising more performance improvements yet to come, Meta shared a bit on how outcomes have been affected by GEM thus far:
“Since launching GEM earlier this year, GEM’s launch across Facebook and Instagram has delivered a 5% increase in ad conversions on Instagram and a 3% increase in ad conversions on Facebook Feed in Q2.
In Q3, we made improvements to GEM’s model architecture that doubled the performance benefit we get from adding a given amount of data and compute.”
Read the deep dive. (November 10)
On Meta’s Q3 earnings call with Wall Street in late October, Meta CFO Susan Li offered a bit more color on GEM: “So we don’t use our larger model architectures like GEM for inference because their size and complexity would make it too cost prohibitive. The way that we drive performance from those models is by using them to transfer knowledge to smaller, lightweight models that are used at runtime.”
Other Meta ad models:
- Meta Andromeda (retrieval engine): Supercharging Advantage+ automation with the next-gen personalized ads retrieval engine (December 2024) – Meta
- (Meta Lattice, i.e. optimization) New AI advancements drive Meta’s ads system performance and efficiency (May 2023) – Meta
From tipsheet: Why publish this GEM “whitepaper” now? I’ll guess two audiences are being targeted by Meta: performance marketers and Wall Street. In so many words, the company is saying, “This is why you invest in Meta. Nobody else has this at our scale.”
SELL-SIDE
SSP PubMatic on AdCP & search
Sell-side platform (SSP) PubMatic reported Q3 2025 revenues on Monday that surpassed analyst predictions. From the release, “Revenue in the third quarter of 2025 was $68.0 million, compared to $71.8 million in the same period of 2024. Q3 2024 included $5.0 million in revenue from political advertising.” Read more. (November 10)
On the earnings call with Wall Street analysts, CEO Rajeev Goel commented on the limited impact of AI chatbots on search traffic among PubMatic’s publisher clients:
“We’ve seen, I would say, a fairly limited impact. I would say just stepping back, we believe the exposure overall in our business is limited to a single-digit percentage of revenue if there was zero search traffic going to our publishers, and we took no steps to mitigate it. That’s probably a high-end kind of watermark of potential impact. We shared in our comments today that roughly 60% of the impressions that we are processing are for CTV and mobile app. Of course, those are unaffected by AI search. Now, of the remaining business, which is browser-based where search is relevant, industry data indicates that search referral traffic is roughly 15%, with the rest of publisher’s traffic coming from either social or direct navigation.”
Read the earnings transcript. (November 10)
Ad Context Protocol (AdCP) is a big part of PubMatic’s future plans according to Mr. Goel. He said on the call:
“…at the transaction layer, we’re preparing for the next major step, Agentic AI, where advertisers and publishers’ AI agents will be able to transact directly through our infrastructure.
We are a co-founder of the newly established Ad Context Protocol, or AdCP, alongside partners like Yahoo, LG Ad Solutions, and Raptive, and the first to publish a model context protocol specification for agent-to-agent communication in the programmatic industry.”
OPINION
On chatbots and ads: Eric Seufert
Mobile Dev Memo analyst Eric Seufert makes his case on why OpenAI’s ChatGPT must incorporate a digital advertising-based revenue model rather than go the affiliate marketing route.
Mr. Seufert writes:
“I believe that personalized advertising will be the model that OpenAI ultimately adopts as the primary monetization scheme for ChatGPT. This is important to consider precisely because OpenAI serves as an avatar for the commercial prospects and general sustainability of the artificial intelligence market, to the extent that the company’s infrastructure commitments are now viewed as systemic economic vulnerabilities. For this reason, the business model that OpenAI adopts to achieve sustainable growth is of critical importance, not just for its own prospects, but also for those of the overall consumer-facing AI category.”
Read: “Affiliate links, personalized ads, and chatbot revenue optimization” (November 11) – Mobile Dev Memo
From tipsheet: OpenAI likely knows it must tread lightly with its revenue plans for the chatbot even if it’s successful for the chatbot. If there’s any hint a chatbot revenue model resulted in the collapse of a vertical or industry, it might give “AI” —and the company— a bad name as well as incur the wrath of governments.
OPINION
On chatbots and ads: Andrew Lipsman
Though he has argued against the viability of agentic commerce, analyst Andrew Lipsman still believes commerce is coming to the chatbot -and with major implications for advertising.
He argued in the latest edition of the Marketecture newsletter:
“While marketers grapple with the implications of AI search on SEO, [OpenAI CEO Sam Altman] is telling us in more ways than one that he already has an ad model in mind—and it’s feeds, not leads.
There are significant ramifications for retail media networks (RMNs). Because while the AI industrial complex ratchets up the rhetoric around agentic commerce’s potential to incinerate ecommerce traffic, retailers’ attention has been diverted from what’s really at stake: off-site advertising.
If RMNs accede their first-party data to ChatGPT, they leave themselves vulnerable to disintermediation. Especially because advertisers believe off-site retail media is too expensive due to retailer data costs…”
Mr. Lipsman offers advice to retail media networks (RMNs) beginning with “simply hold tight.”
Read: “Open AI’s Ad Model Will Be Feeds, Not Leads” – Andrew Lipsman on Marketecture (November 10)
MORE
- AI agent developers have become adland’s in-demand role (November 10) – Digiday
- Matthew McConaughey, Michael Caine Ink Deals for AI-Generated Versions of Their Voices With ElevenLabs (November 11) – Variety
- Index Exchange Sues Google Over Ad Auctions Practices, Joining Ballooning Coalition of Publishers, SSPs (November 10) – Adweek (subscription)