User-generated recommendation company Bazaarvoice announced a new AI visibility solution yesterday with GEO measurement support from Bluefish, Evertune and Profound.
The opportunity, according to Bazaarvoice:
“AI is fast becoming the new storefront. But most brands’ hardest-won proof of trust, from customer reviews to photos and ratings, is invisible to it. That’s because most review platforms render this content through JavaScript widgets, a format most AI crawlers skim past entirely. As shopping increasingly starts with an AI-generated answer rather than a search results page, content that AI can’t see can’t be surfaced.
The AI Visibility package is built to close this gap.”
Read: Bazaarvoice introduces AI Visibility package to prime brand content and UGC for AI recommendation (Sept 1) – press release
From tipsheet: Bazaarvoice is supplying the content infrastructure; Bluefish, Evertune and Profound are the independent measurement layer tracking whether that content appears in AI answers. It’s a different role for firms that are often positioned as AI visibility solutions themselves.
Related: AI-generated reviews — “Google Ads AI-Generated From The Store Rating Reviews Label” (Sept 1) – SE Roundtable
MEASUREMENT
Job: OpenAI expands ads measurement team
OpenAI is hiring a “Marketing Scientist” for its Ads Marketing Science function with a mandate to help “prove and improve advertisers’ media performance.”
The role will work with advertisers, agencies and partners across “verification, attribution, incrementality, brand, marketing mix modeling (MMM), and scalable advertiser reporting.” OpenAI is also looking for experience building lift and incrementality programs and working with third-party measurement providers.
See: Marketing Scientist – OpenAI Careers
From tipsheet: The inclusion of incrementality and MMM is particularly notable. OpenAI appears to be planning for advertisers to measure both causal lift and advertising’s contribution within the broader marketing mix.
AGENTS
Opinion: Hey agent, who do you work for really?
A recent change in how Google Ads target-based bidding works on budget-limited campaigns is the subject of an op-ed on AdExchanger by Toni Poulain, SVP of Media Operations, Goodway Group.
She believes the change exposes a larger problem as platforms push advertisers toward autonomous media buying and agentic workflows.
Poulain argues:
“The more platforms automate, the more advertisers need people empowered to challenge what the machine recommends. Automate the keystrokes. Do not automate away the skepticism….
No agency can claim to be the advertiser’s independent agent while relaying platform recommendations, optimizing to platform-reported conversions or attaching its own economics to higher spend…”
Read: Google’s Bidding Change Proves Agentic Advertising Has An Agency Problem (Sept 1) – Toni Poulain, Goodway Group on AdExchanger
From tipsheet: Whose interests does the agent represent? Automation does not eliminate conflicting incentives.
PLATFORMS
Reaction: FTC sues Amazon over ad practices
Industry reaction is coming in after the Federal Trade Commission sued Amazon and alleged that the company secretly manipulated its ad auctions to make advertisers pay higher prices.
Eric Seufert, analyst, Mobile Dev Memo on X (Aug 31):
“It may be reasonable to conclude that bid floors are hostile to advertisers, even when the practice is disclosed; I don’t have a well-considered view on that. But I’m having a hard time understanding how the FTC can demonstrate that they result in costs being passed along to consumers at the SKU level.”
Param Vir Singh, Associate Dean, Carnegie Mellon on LinkedIn (Sept 1):
“There is nothing inherently wrong with a platform trying to extract more value from an auction.
The controversy is whether advertisers believed they were bidding under one pricing mechanism while the platform was effectively using another. That distinction matters because bidding behavior depends on the rules. If I believe I am in a true second-price auction, I have an incentive to bid close to my actual value. If the mechanism can instead push my payment toward my maximum bid, my optimal bidding strategy changes.”
Tim Vanderhook, CEO, Viant on LinkedIn (Sept 1):
“Looks like Amazon copied Google’s playbook.
Play #1 at Google was Reserve Price Optimization. A second-price auction with an artificial bid pushing your price up. Amazon calls this a “soft reserve.” The FTC and 22 states just sued them for running it for years while still calling the auction second price.
Play #2 at Google was Project Bernanke. Advertiser pays one price, the publisher gets a lower price and the platform keeps the spread.
If they copied Play #1, advertisers should be asking if they copied Play #2.”
More: “WTF is the Amazon ad surcharge lawsuit?” (Sept 1) – Digiday (subscription)
From tipsheet: The early debate is separating into two questions: Did Amazon’s auction design harm advertisers and consumers? And did advertisers understand the rules of the auction they were bidding into? The second may matter well beyond this lawsuit as machines increasingly make bidding decisions on advertisers’ behalf.
LLMs & CHATBOTS
Developments
- Claude: Fable 5.1 and Mythos 5.1 (Sept 1) – Anthropic
- AI token prices are hitting new record lows (Sept 1) – CNBC
- Introducing Muse Voice Transcribe (Sept 1) – Meta
MEASUREMENT
Survey: Room to grow in AI search tools
B2B marketing agency Gilroy analyzed 555,973 online opinions from UK B2B marketing and revenue leaders and found 83% say better tools for monitoring brand representation in AI search would help improve performance across AI-driven buyer journeys.
The research also found 68% don’t really monitor how their organization appears in tools such as ChatGPT or Perplexity, while 67% say buyers typically research solutions through AI search before engaging with sales.
Read: The B2B buyer journey now starts in AI search (Sept 1) – The Drum
From tipsheet: 83%? Whoa.
Glass-half-full take: Startup leaders in the AI visibility space have room to run.
AGENCIES
AI reshapes WPP?
Anonymous sources tell the FT that WPP is laying off more people — up to 1,000 this time. WPP has eliminated approximately 11,000 jobs since the beginning of 2025, reports the FT.
In its most recent H1 2026 earnings report to the U.S. Securities and Exchange Commission, WPP said, “The average number of people in the Group in the first half was 97,490 compared to 105,958 in H1 2025.”
Read: “WPP cutting up to 1,000 more jobs as AI reshapes advertising industry” (Sept 1) – Financial Times (subscription)
From tipsheet: In the bigger scheme of WPP, this is a little over 1% of the workforce. So, is the latest round of layoffs AI, bloat or a performance-based culling of the herd?
Related: Small agencies benefit from fewer layers—how many you can really eliminate (Sept 1) – Ad Age
GOVERNANCE
AI disclosure: Structuring the provenance data
On Monday, Meta announced that the Instagram app would have new, stricter guidelines regarding AI-generated images as AI slop swarms the app.
From the announcement:
- “Creators who don’t appropriately label their AI-generated profiles may see limits to their profile’s reach. Unlabeled profiles that we detect will be notified and can check Account Status to see if they’re no longer eligible to appear in recommendations.”
Read: Bringing More Transparency to AI-Generated Profiles on Instagram (Aug 31) – Instagram’s Creators blog
Creative provenance is becoming infrastructure:
- Google announced its provenance disclosure “How this ad was made” across Search, YouTube and Discover (tipsheet, July 10).
- Meta issued another update to its AI disclosure policy, too (tipsheet, July 9).
- IAB is working on a “Transparency and Disclosure” framework (tipsheet, Aug. 19) for disclosure and ad creative’s underlying provenance.
From tipsheet: Provenance is becoming a control layer, too, meaning information about how an asset was created becomes structured data that platforms can act on, determining how creative is labeled, distributed or recommended. In an AI world, disclosure becomes something a machine can use.
MORE
- Ad lovers: “Ad Market Continues World Cup-Related Surge, Expands 11.9% In July” (Aug 31) – MediaPost
- Ad haters: “Introducing Ad Blocker for Firefox on iOS: More Control, Fewer Distractions” (Sept 1) – Mozilla blog
- Distilling expertise: Inside Microsoft’s marketing team: Scaling expertise with AI (Sept 1) – Microsoft Azure blog
- Executable brief: India’s “Voiro Expands Its Intelligent Media OS with Agentic AI and Programmatic Intelligence” (Sept 1) – press release