No in-house AI for Apple. It’s chosen a partner instead.
From a joint statement yesterday by Google and Apple yesterday:
“Apple and Google have entered into a multi-year collaboration under which the next generation of Apple Foundation Models will be based on Google’s Gemini models and cloud technology. These models will help power future Apple Intelligence features, including a more personalized Siri coming this year.
After careful evaluation, Apple determined that Google’s Al technology provides the most capable foundation for Apple Foundation Models and is excited about the innovative new experiences it will unlock for Apple users. Apple Intelligence will continue to run on Apple devices and Private Cloud Compute, while maintaining Apple’s industry-leading privacy standards.”
Read: Joint statement from Google and Apple (January 12) – Google on The Keyword blog
This is not unexpected news (rumored in June and August).
Google makes sense an AI partner considering its leadership position in AI and also Apple/Google’s partnership in search: Google Search is the preferred search engine in Apple’s Safari browser and generates billions in revenue for Apple.
More: “Apple picks Google’s Gemini to run AI-powered Siri coming this year” (January 12) – CNBC
From tipsheet: Though relatively insignificant from a “walled garden” perspective, I wonder if Apple’s estimated $10 billion app store ad business is at some point destined for a refresh courtesy of Google AI.
LLMS & CHATBOTS
Developments
- Cowork: Claude Code for the rest of your work (January 12) – Anthropic
- Meta launches Meta Compute to build AI infrastructure (January 12) – Meta CEO Mark Zuckerberg on Facebook (FB login required)
- OpenAI Agrees to Buy AI Healthcare App For About $100 Million (January 12) – The Information (subscription)
LLMs & CHATBOTS
TBPN: Stratechery’s Thompson on ChatGPT ads
“Analyst Ben Thompson says the fact that OpenAI hasn’t already put ads in GPT is a poor decision that ‘risks the entire company.’
‘They could have launched the world’s crappiest ads in 2023. By today, in 2026, the ads would be good. They’d be making money, and people wouldn’t rebel against it.’
‘Now, they’re going to have to launch ads, they’re going to suck, and people are going to be like “This sucks, I’ll just go to Gemini.”’”
See a clip from Ben Thompson’s TBPN interview on X. (January 12)
Hear the entire TBPN interview of Thompson on Apple Podcasts app [begins 29:54]. (January 12)
TECH
New GEO platform gets funding
A new startup called Emberos, which is focused on the busy generative engine optimization (GEO/AEO) category, announced $1.2 million in pre-seed funding from a single unnamed investor yesterday.
Adweek’s Trishla Ostwal talked to the startup’s CEO Justin Inman about his company’s plans.
Mr. Inman is a former Googler who more recently spent nearly 3 years in enterprise sales at Cardlytics, according to his LinkedIn profile.
On how Emberos will differentiate in the GEO space, Ostwal wrote, in part:
“Rather than reporting outcomes after the fact, Inman claims the platform predicts how brands will appear in AI engines, and pushes recommended fixes directly into existing marketing workflows.
‘A lot of our competitors are looking at it from a reactive standpoint—telling you what has already happened. We have a predictive capability, and then we give you what we call fix packs,’ Inman said. ‘I want to demystify that black box and make it more approachable and automated—not just for the executive level, but also the practitioner level.’”
Read: “AI Visibility Startup Emberos Raises $1.2M in Pre-Seed Funding” (January 12) – Adweek
More: “Startup Emberos Shows AI ‘Share of Prompt’ Can Predict Sales With Wicked: For Good Case Study” (November 2025) – press release
From tipsheet: On the company’s home page, Emberos positions itself as “the Operating System for AI Brand Visibility.”
“Brand Visibility” feels like an important feature of a larger, cross-functional marketing platform. Perhaps that is the end game for many of these GEO/AEO companies.
PROTOCOLS
Explaining AdCP governance
Matthew Egol, one of the leaders of the new governance organization focused on supporting Ad Context Protocol (AdCP), announced on LinkedIn a new Substack post co-written by his AgenticAdvertising[dot]org (AAO) colleague Randall Rothenberg.
In the comprehensive post, Messrs. Egol and Rothenberg discussed their broader goal of “accelerating the builder-led agentic advertising movement.” They bring together everything from historical context to highlighting voices from AAO’s current membership to explain its agentic advertising vision.
The duo also aimed to clarify:
“To give away the plot, we do not envision AgenticAdvertising[dot]org (AAO) as an industry association because agentic advertising – what we just described – is not an industry, but rather a profession, a discipline, and a practice built on a paradigm. AAO embraces the mental model that advertising succeeds by encompassing the totality of consumer experiences, not just formatted spaces, time slots, and impressions.”
On “who” belongs in AAO:
“If you’re a Chief Marketing Officer, Chief Revenue Officer, Chief Experience Officer, or other C-suite leader accountable for growth and wondering how agentic marketing reshapes your organization, AAO is your community. If you’re building a DTC company competing against larger players (or if you’re a giant brand grappling against these nimble upstarts) AAO is your resource. If you’re an agency leader navigating rebundling, AAO is your forum. If you’re a publisher reimagining your business model, AAO is your coalition. If you’re an AI marketing specialist establishing best practices, AAO is your professional home.”
Read: “Welcome to the Cre(ai)tive Economy” (January 12) – “Winning Experience” Substack (January 12)
From tipsheet: FYI, my brief summary doesn’t do the post justice. Lots in there.
Also, as the organization promised on the launch of AAO, there is an “entry point” for individuals not just companies. See the AgenticAdvertising[dot]org’s website for membership pricing.
Regarding AAO saying it is “not a trade association,” this may be tied to recent industry friction over agentic protocol development related to AdCP and IAB Tech Lab’s ARTF initiative. AAO isn’t looking to do battle with the much larger IAB trade association… appears to be the message.
LLMs & CHATBOTS
OpenAI’s ChatGPT versus Google’s Gemini
eMarketer chief content officer Vladimir Hanzlik delivered another eMarketer chart focused on the fast-growing world of chatbots yesterday. He and his team tried to understand who’s winning between OpenAI’s ChatGPT and Google’s Gemini.
At first, Mr. Hanzlik explained:
“Until now, it’s not been possible to compare the adoption of ChatGPT and Gemini fairly. ChatGPT shares weekly active users (800 million in October ’25 per Sam Altman) while Gemini reports monthly active users (650 million in October ‘25 per Alphabet’s Q3 earnings call).
But with the recent data by Sensor Tower on the monthly active users of mobile apps, we can pull together a decent back-of-the-envelope estimate of ChatGPT’s monthly active users in October ’25 of about 1.2 billion, which is 88% more than Gemini’s 650 million monthly active users.”
Later, eMarketer analyst Nate Elliott added in the LinkedIn comments on the post that his firm’s analysis still believes Google will overtake OpenAI’s ChatGPT in consumer AI adoption in 2026 given Google’s AI chatbot proxies — AI Overviews and AI Mode — and Gemini.
Read more on LinkedIn. (January 12)
TECH
Viant pitching autonomy and no ‘black box’
Viant CEO Tim Vanderhook and COO Chris Vanderhook joined Ari Paparo on the latest episode of the Marketecture podcast to discuss Viant’s new autonomous advertising product, “Outcomes, powered by Lattice Brain AI”, which was launched at CES last week.
More: “Viant Launches ‘Outcomes’ the First Fully Autonomous Open Internet Ad Product” (January 5) – Viant
Lattice Brain AI underpins “AI decisioning” in the Outcomes platform and aims to fit within the competitive set of automated ad platforms such as Google’s Performance Max and Meta’s Advantage+, according to CEO Vanderhook.
Vanderhook also emphasized a key difference between automated and human-operated ad platforms:
“Once you implement an autonomous ‘brain,’ you remove decision latency. And latency is where campaign performance leaks. That is the line of thought on why AI wins: there is no decision latency or ‘wait ‘til tomorrow’… this thing just makes changes constantly.”
Still another differentiator according to CEO Vanderhook is no “black box” unlike Performance Max or Advantage+. With Viant’s Outcomes, Vanderhook said you get “full transparency, but no control.”
Hear more on Apple Podcasts app. (January 12)
From tipsheet: The Vanderhooks are the founders of “old school” ad network Specific Media which ultimately re-branded into Viant along with other products in 2015. Today, publicly-traded Viant, valued in the public markets at ~$700 million, and positions itself as a CTV and AI-powered programmatic advertising platform.
ANALYSIS
Regarding Google’s agentic commerce vision
“Ultimately, my interpretation of these tools is that Google sees AI enablement as a source of optimization for its existing advertising business — the continuation of its ‘gambit’ — and not as a means of completely overhauling that business. Further to that point: if it can allow retailers to offer on-site agentic shopping capabilities, it can reduce the appeal of integrating with independent chatbots for discovery, further emphasizing the need to use ad budgets for that purpose.”
Read: “Google launches Universal Commerce Protocol, ads in AI Mode” (January 12) – Analyst Eric Seufert on Mobile Dev Memo (subscription)
AFFILIATE MARKETING
Rakuten brings AI to affiliate marketing
Affiliate marketing company Rakuten Advertising is embracing AI for its product set.
The company announced the launch of “Innovation Labs” yesterday which it described as “a product collaboration hub dedicated to accelerating the development and application of AI to generate greater ad spend efficiency to target and engage customers…”
In the announcement, Rakuten CEO Nick Stamos said:
“This initiative is a critical step in how Rakuten Advertising contributes to the broader Rakuten Group’s ‘Triple 20‘ commitment to using AI to drive 20% greater efficiency for our employees, our internal teams, and most importantly, our clients. Innovation Labs will shorten the build–measure–learn cycle of AI transformation, helping us fast-track future investment in capabilities that improve outcomes for our advertisers and publisher partners.”
Rakuten Advertising explained how Innovation Labs will impact its affiliate marketing customers initially:
- “For Advertisers: AI Recommendations will holistically analyze multidimensional advertiser attributes to determine new types of publishers. This involves analyzing profiles, products, and other characteristics to identify publishers with demonstrated success in similar campaigns, and also identifying new categories of publishers that will maximize growth.”
- “For Publishers: AI-driven Product Recommendations reduces research time and enhances monetization for publishers by instantly identifying emerging trends from partnered and non-partnered advertisers, then curating them for a targeted audience and narrowing options with brand and category filters.”
Read: “Rakuten Advertising Unveils Innovation Labs to Accelerate AI Innovation in Affiliate Marketing” (January 12) – press release
Related: “This morning Rakuten ejected [PayPal’s] Honey from its network. Arguably its hand was forced — such strong evidence of Honey not just failing to ‘stand down’ (as policy required), but hiding from testers…” (January 12) – consultant Ben Edelman on LinkedIn
SELL-SIDE
Report: Publishing’s referral traffic woes
“Publishers expect traffic from search engines to decline by more than 40% over the next three years – not quite ‘Google Zero’ but a substantial impact none the less. Data sourced for this report from analytics provider Chartbeat shows that aggregate traffic to hundreds of news sites from Google search has already started to dip, with publishers that rely on lifestyle content saying they have been particularly affected by the roll out of Google’s AI overviews. This comes after substantial falls in referral traffic to news sites from Facebook (-43%) and X, formerly Twitter (-46%) over the last three years.”
Read: Journalism, media, and technology trends and predictions 2026 (January 12)- Reuters Institute
Summary: News publishers expect search traffic to drop 43% by 2029 (January 12) – Search Engine Land
MEASUREMENT
Modernizing geographic experimentation
Former agency professional and consultant Nate Woodman examines the future of media measurement in a self-published article on LinkedIn yesterday.
He thinks “geo” – as in “geographic” – is going to matter more than ever and explains:
“…The future of measurement is not choosing between attribution, econometric modeling, or experimentation. It is modernizing geo experimentation so it can serve as a durable, system-wide source of causal reference.
By combining continuous geo experiments with spatio-temporal embeddings, organizations can improve experimental rigor, enable cross-platform learning, and align otherwise isolated measurement systems around a shared understanding of incrementality.
This methodology is intentionally generic. Its value lies in its ability to evolve alongside modeling techniques, data availability, and platform constraints, while preserving the central role of causality in decision-making.”
Read: “Modernizing Geo Experiments with Spatio-Temporal Embeddings” (January 12) – Nate Woodman on LinkedIn
PEOPLE MOVES
Executive departure
-
Reddit CRO Mike Romoff announced his departure from the company yesterday: “As my time at Reddit comes to an end, I’m incredibly proud of what our team accomplished together…” (January 12) – Mike Romoff on LinkedIn
MORE
- How Google, Shopify Partnership Could Spur Agentic Commerce Boom (January 12) – Investor’s Business Daily
- OpenX Unveils OpenXBuild: A New Software Suite for Building Better Advertising Solutions (January 6) – press release
- Media buyers say Yahoo’s DSP is one to watch (January 12) – Digiday (subscription)
- Treasure Data Launches Marketing Super Agent: The First Enterprise AI That Operates Like a Marketing Department (January 12) – press release