Moving beyond its chatbot optimization roots, Bluefish introduced Bluefish Agentic Ads yesterday, adding paid media capabilities to its agentic marketing platform (AMP).
A press release said: “The launch marks a significant evolution of the Bluefish AMP, extending its foundation in organic Answer Engine Optimization (AEO) to help enterprises move toward orchestrating every lever that shapes their brand’s performance in AI.” Enter the paid ad.
Read: “Bluefish Expands Agentic Marketing Platform Into Paid Media to Unite the AI Marketing Funnel” (Oct. 27) – Bluefish
From tipsheet: Bluefish’s pitch is that what AI says about a brand should inform where it spends: which audiences to target, what messages to use and how to allocate budgets.
Bluefish CEO Alex Sherman answered follow-up questions from tipsheet by email yesterday.
Responses have been edited for length and clarity.
tipsheet: How is Bluefish thinking about paid media and its role in agentic marketing?
Alex Sherman: How an AI platform describes a brand, which competitors it recommends and how it ultimately shapes the AI customer journey — all inform where and how to allocate paid media dollars.
For Bluefish customers, this means that brands can now use their AI performance signal to optimize their paid media strategies across those channels — from identifying targeting opportunities, optimizing messaging and ad types, and allocating budgets. With Bluefish Agentic Ads capabilities, marketers now have a unified way to track and manage the paid and organic activities that influence their AI customer journey.
tipsheet: Is paid media becoming an essential component of generative engine optimization (GEO)?
Alex Sherman: We actually see AEO as a small part of a much larger transition that is underway. Simply put, AI has introduced an entirely new agentic customer journey, fundamentally changing how consumers discover and choose brands. For brands, this isn’t simply about adding paid media to AEO — it is about rethinking how marketing should be optimized holistically around this new customer journey.
tipsheet: Some competitors in the GEO space have moved into campaign activation. Can you envision Bluefish doing the same, or do you see greater value in remaining an intelligence and orchestration layer?
Alex Sherman: Bluefish already operates well beyond intelligence and orchestration — over half of our customers use Bluefish to actively optimize and execute their agentic marketing strategies across channels.
Our AI Optimization Suite includes a broad range of execution levers to improve how brands are represented and recommended in AI, and our Agentic Campaigns tool enables marketers to launch AI optimization campaigns that run across marketing functions and channels. Many of these channels, like paid, are already mature and have a robust ecosystem of vendors that our customers use every day — in these cases, we tend to partner strategically and have done so with a number of category-leading platforms.
tipsheet: Where do you stand on advertising designed specifically for — and visible only to — AI agents?
Alex Sherman: Something we are tracking closely. We believe agent-directed advertising will become an important area of experimentation as AI agents take on a greater role in making purchases on behalf of consumers. That said, obviously advertising to an agent is fundamentally different from advertising to a person. An agent may evaluate structured product information, pricing, availability and other signals against a consumer’s stated preferences, rather than respond to traditional creative or an emotionally driven message.
COMMERCE MEDIA
ADstruc co-founder targets agentic shopping
John Laramie, whose startup ADstruc was sold to Vistar Media in 2024, is getting busy in the agentic shopping category.
Yesterday, Laramie outlined CartHappy, a Chrome browser extension for agentic shopping. He explained, “We check more than 400 million grocery prices a day and match them product by product, so a cart can be split across stores when that lowers the total.”
Read more on LinkedIn. (Oct. 27)
CartHappy’s pitch includes working “across every store you shop instead of living inside any one of them, so for the first time the full picture is in one place — every price, every coupon, every week.”
Asked by tipsheet whether CartHappy uses Universal Commerce Protocol (UCP), Laramie said: “We are not — we’ve built everything internally … UCP just wasn’t there yet to the level we needed it.”
See: CartHappy – Chrome extension store
From tipsheet: Honey for groceries. Will it be sticky?
PLATFORMS
Jeff Green: ‘The case for the Open Internet’
In a new op-ed, The Trade Desk CEO Jeff Green returns to his company’s publication, The Current, to argue that AI makes the open internet more important.
The essay also kicks off a revamped The Current. Green said on LinkedIn that the publication will carry multiple op-eds each week from industry leaders, most of them outside The Trade Desk.
Green argues that AI threatens to concentrate control over information — and the money that funds its creation — in even fewer hands. He writes:
“In this dystopia, media companies are anemic and paying rent to a few tech companies who became the gatekeepers for the marketplace of ideas—first by indexing the world’s information and then by just taking it.”
Advertisers can counter that concentration by directing more spending toward independent, premium media, he says.
Read: “The case for the Open Internet.” (Oct. 27) – The Current
From tipsheet: Green makes the societal case for supporting independent media while advertisers still need the economic case for shifting their budgets.
CONNECTED TV
FreeWheel acknowledges its future
Video ad technology company FreeWheel published a blog post Monday confirming tipsheet’s exclusive reporting Friday that it would join NBCUniversal post-Comcast split.
Its self-serve offspring, Universal Ads, will too.
Read: “Building for Premium Video’s Next Chapter” (Oct. 26) – FreeWheel blog
(h/t AdExchanger)
SEARCH
Conversational ads are different
On LinkedIn Monday, Vijaye Raj, CTO of Applications at OpenAI – the guy in charge of creating the ChatGPT Ads product – explained why ads in a conversation need a different system from search ads.
“Building ads for a conversation is different from building ads for a search query. Intent evolves across turns, so we built an ad stack that can follow the conversation while keeping ads separate from the answer.
Also latency matters…”
Read more. (Oct. 26)
Raj was part of Meta’s early ads team.
Read: “How we built ChatGPT’s novel ad stack for user trust and relevance” (Oct. 21) – OpenAI
CONNECTED TV
Fox-Roku may be closer
An October 21 report from antitrust specialist The Capitol Forum says the $22 billion Fox–Roku acquisition is moving toward possible Department of Justice clearance.
According to its reporting, DOJ staff is wrapping up its investigation, with no indication so far that department leadership intends to challenge the transaction.
Meanwhile, White House adviser Peter Navarro remains opposed to the acquisition. Read more on LinkedIn. (Oct. 22)
Read: “Fox/Roku: Streaming Deal Moves Toward DOJ Clearance” (Oct. 21) – The Capitol Forum (subscription)
LLMs & CHATBOTS
Briefings first, delegation later, says Kovva
For media buyers using Kovva, handing work to agents starts with checking their work.
Tanja Mimica, co-founder and COO of Kovva, discusses her agentic AI startup on the AdExchanger Talks podcast.
Previewing the episode, AdExchanger’s Allison Schiff reports:
“Buyers typically start with a daily or weekly briefing sent to email, Slack or Teams that recaps the past 24 hours in their accounts. They can cross-check that with what they’re seeing in the platforms, ask the agent follow-up questions and, once it proves itself, start scheduling recurring tasks and handing off more complex workflows, a shift that takes at least three to four weeks.”
Read more — and listen on YouTube. (Oct. 27)
From tipsheet: Can Kovva turn that gradual handoff into a moat? The race is on.
COMMERCE MEDIA
Amazon: AI advertising should feel like help
Fresh from her appearance at the Amazon Beauty and Fashion Summit, Amazon conversational AI and monetization executive Ali Peikon Solomon shares highlights from the event:
“A few takeaways shared:
1. Brands that lean into AI-powered shopping experiences early are building a real competitive advantage.
2. The best AI advertising doesn’t feel like advertising. It feels like help.
3. We’re still in the early innings, and the brands willing to experiment and learn will define what comes next!”
Read more. (Oct. 27)
From tipsheet: Buy now.
PLATFORMS
Debate: Brand-owned versus platform AI
What can a brand-owned model demonstrably achieve that the platforms’ increasingly customizable AI cannot?
On his personal blog, Chalice AI co-founder and CEO Adam Heimlich makes his case — and his startup’s. He writes:
“The claim here is not that pooled models are inaccurate. The claim is they converge, and that convergence is wrong for strategy. It’s getting worse with platform AI, not better. The chatbot’s spike came from more tuning, not less. The harder a shared system optimizes toward a shared objective, the more uniform its outputs across the advertisers inside it.”
Read: “Platforms Pick 7.” (Oct. 27) – Adam loves ad tech blog
LLMs & CHATBOTS
Developments
-
- Genie 4: A new generation of world simulation (Oct. 27) – Google DeepMind
- ChatGPT’s expanding world of work (Oct. 26) – OpenAI
- Apple is adding an AI chatbot to Pages, Keynote, and Numbers (Oct. 27) – The Verge (subscription)
MORE
-
- AI gets CTV: “How a mid-size brand is testing AI-powered connected TV buying” using Kargo (Oct. 27) – Ad Age
- Meta wants CTV: “Instagram Goes Horizontal: Tests Long-Form Content On Mobile, TV” (Oct. 26) – MediaPost
- SEO alert: “Poisoned SERPs: An update on Google ranking data in Semrush” (Oct. 27) – Adobe’s Semrush