Your weekly guide on everything AI marketing. Covering everything from AI marketing news, tips, deep dives, events, podcasts, jobs and much more. Never miss a beat with our 5-minute newsletter.
In today’s email:
Yahoo opens its DSP to 23 third-party AI agents
LinkedIn and Adobe launch free AI training as job ads demand it
MoEngage buys Aampe to put an AI agent behind every customer
Plus: a 30-minute inbox-automation workflow, four fresh tools, a podcast on the agentic ad era, and why too many AI pilots is the new marketing trap. And, AI Marketers is officially live!
The Top 3 Stories
1. Yahoo opens its ad platform to outside AI agents
Yahoo’s DSP launched an Agent Network connecting advertisers to AI agents from 23 ad-tech partners across creative, measurement and targeting in one authenticated environment. Its “Yours, Mine and Ours” approach lets buyers bring their own AI, use Yahoo’s, or combine both via open protocols. Yahoo is betting openness beats the closed agent stacks rivals are pushing.
Why it matters: As agents take over campaign work, open platforms are easier to audit and trust than proprietary black boxes. eMarketer
2. LinkedIn and Adobe give marketers free AI training
LinkedIn and Adobe launched AI Essentials for Marketers, four role-based learning paths on LinkedIn Learning covering AI content, targeting and agentic workflows, free and in 47 languages. Each takes two to three hours and earns a profile certificate. The trigger is stark: AI-literacy job postings on LinkedIn are up 113% year on year, while only 4% of marketers list AI skills.
Why it matters: The skills gap is measurable now, and “can you direct AI” is becoming a hard hiring filter for marketing roles. Social Media Today
3. MoEngage buys Aampe to give every customer its own AI agent
Customer-engagement firm MoEngage acquired San Francisco startup Aampe, whose software provisions a dedicated AI agent for each individual customer that decides what to say, when, how often and on which channel. The bet is that personalisation shifts from rules and segments to millions of per-person agents acting in real time. MoEngage frames it as the end of the segment-and-blast era.
Why it matters: If personalisation goes per-customer-agent, the old segment-and-send playbook quickly looks blunt. TechCrunch
More trending AI marketing news from last week
Snap spun off its generative AI video team into a new company, Dotmo.
Hootsuite rebuilt around Wisdom, a social-first AI agent, and launched Social OS.
US generative-engine-optimisation market is forecast to hit $365m in 2026.
Marketing leaders say AI agent adoption is outrunning marketing readiness.
“AgentOps” is being floated as marketing's next discipline, after DevOps.
AI Marketers is now officially… LIVE!
If you've not claimed your account yet, start with the AI Audit. It’s also completely free!
After a short survey, you get back a maturity score across your stack, workflows and team, plus a tailored playbook and a 90-day roadmap that you can implement from the get-go.
It's one part of the platform: tool directory, jobs board and events sit alongside it, and your portal saves whatever you want to come back to. Your account's still reserved from last week. Claim your account and take the audit today:
Quote of the week
The role of the CMO is going to become more about orchestration: how do you make a great idea come to life, not just in one channel, but throughout the ecosystem.
Trending AI tools for marketers
Here are four tools worth a look this week. Quick, practical upgrades to your AI marketing arsenal:
Elentaria: AI operator that scores 22 channels and ships your weekly GTM plan.
SellerClaw: A team of AI agents that runs your stores across channels.
CiteLens: Tracks how often AI engines cite your brand versus rivals.
Upstream: An inbox built for humans and agents to work side by side.
TV, podcasts & streaming
The AI Advertising Podcast (StackAdapt): inside the agentic ad era
The show digs into how AI agents are moving out of pilots and into live ad workflows, and what that does to the people running them. The sharpest stat: 90% of marketers say they use AI agents somewhere, but only 23% have them in full production. The hosts argue the bottleneck is no longer the technology but the trust and process to let agents run. For marketers, that gap is the whole story - listen to it here.
AI training
How to turn your inbox into an AI-run workflow using Upstream
The overview: Upstream connects to your existing Gmail and puts AI agents to work triaging, drafting and chasing replies, so you clear the inbox in minutes rather than hours.
Step-by-step:
Connect your Gmail
Link your existing address; Upstream syncs in place rather than making you move inboxes. Connect more than one account if you run several.Let the agents triage first
On first run, agents sort signal from noise so you see only what needs you. Spend five minutes correcting miscategorised threads to train it faster.Review the auto-drafts
Every thread that needs a reply gets a draft in your voice, adapted to the recipient. Always read before sending; AI tone drifts generic on cold threads.Set follow-up reminders
Tell it to track open loops so prospects and partners do not slip. Good output is a short chase list, not a buried flag.Query across your tools
Ask it to pull receipts, notes or past decisions from connected apps like Calendar and Drive. Treat answers as a fast first draft, then verify anything you act on.
Pro tip: Spend ten minutes refining the writing-style prompt in week one; a sharper voice profile is what makes later drafts genuinely send-ready.
The weekly deep dive
Your AI problem is too many projects, not too few
The fashionable failure of 2026 is not ignoring AI, it is launching so much of it that nothing compounds.
What the data shows
At Fortune’s Brainstorm Tech, executives from West Monroe, Amgen, Carvana and GXO described companies drowning in AI, standing up hundreds of disconnected pilots that are never finished or serve no strategic purpose. As West Monroe’s Brett Greenstein put it, every team claims 300 use cases when only three move the needle. Sprawl, not scarcity, is the constraint.
Why marketing is especially exposed
Marketing runs the most fragmented stack in the business, so every new AI pilot adds another silo of data, billing and logins, each with its own champion and KPIs. Amgen’s Sean Bruich warned the explosion of pilots becomes “an incredible drag on your ability to move quickly,” and that firms are too slow to kill the ones that do not deliver. Busy dashboards, flat outcomes.
What good looks like instead
Carvana’s Dan Gill put it bluntly: getting one thing all the way done beats five things advanced to 20% each. The teams pulling ahead are killing pilots and concentrating on the two or three that moved a number, so each survivor gets the data and attention to compound. Fewer, deeper bets is the winning move.
Our takeaway: Audit your AI projects this quarter and cut the ones with no owner or no metric, because in 2026 the edge comes from finishing a few, not starting many.
📖 Read the full article at Fortune
