In partnership with
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:
Dubai mandates agentic AI across its private sector
Meta's Business AI hits 10 million conversations a week
Anthropic's co-founder calls 60% on self-improving AI by 2028
Plus: 50% of Super Bowl spots to utilise generative AI, get to grips with Hightouch, and why marketers are rebuilding their stack with AI-built apps, not SaaS.
The Top 3 Stories
1. Dubai gives its private sector two years to go agentic
Crown Prince Sheikh Hamdan announced on 4 May that every company in Dubai must adopt agentic AI within two years, backed by funded incubators, training tracks across the Chamber of Commerce, and dedicated investment funds. The brief follows the UAE cabinet's April order that 50% of federal services run through autonomous AI agents by 2028. It is the most aggressive private-sector AI mandate any government has issued.
Why it matters: This is the first time a government has put a fixed deadline on agentic AI; expect every other GCC market to copy the brief. 📰 TheNextWeb
2. Meta's Business AI hits 10 million conversations a week
Meta launched Ads AI Connectors on 29 April, piping live ad accounts straight into ChatGPT, Claude and Codex through MCP, while its in-app Business AI is now brokering 10 million conversations a week between brands and customers. The combination puts a generative interface on both the buyer and the seller side of every Meta ad. Small businesses are getting the bulk of the early lift on the customer-facing side."
Why it matters: Meta has effectively turned ChatGPT into a paid media console; the agencies that miss this beat will be undercut by anyone with a Claude tab open.📰 TechCrunch
3. Anthropic's co-founder gives self-improving AI 60% odds by 2028
Anthropic's co-founder Jack Clark put a 60% probability on AI being able to autonomously train its own successor by the end of 2028. He cites Claude Opus going from a 2.9x training speedup in May 2025 to over 50x in early 2026 as evidence the curve is bending up fast. He also warns that even a 99.9% alignment approach decays exponentially once recursion kicks in.
Why it matters: Your three-year content, SEO and brand plans need a written assumption about what happens if model capability keeps compounding. 📰 Import AI
More trending AI marketing news from last week
Perplexity's Computer agent went live for the enterprise, slotting into Slack and replacing six-figure marketing tool stacks in a weekend.
Companies are now designing themselves around AI agents from day one, with founders shipping product before they hire humans.
Microsoft launched natural-language audience generation, letting advertisers describe a customer in plain English and skip manual segment building.
AI transformation hinges on basics, not bigger models, says HBR, with most pilots stalling on data hygiene and workflow design.
UKTV added AI to its CMO's remit, retitling the role CMAIO and giving one leader the brief across both functions.
ServiceNow projects $30bn in 2030 revenue on the back of its Now Assist AI uplift, naming AI as the single biggest growth driver.
Stop babysitting dashboards. Ship from Slack. Touch grass.
700+ teams have Viktor reading their Google Ads every morning.
Your media team opens Slack at 8am. There's a cross-platform brief in #growth: Google Ads spend vs. ROAS, Meta CPA by campaign, Stripe revenue by channel. Viktor posted it at 6am. Nobody asked for it.
Last week, one team's Viktor caught a spend spike at 2am on a broad match campaign and flagged it in Slack: "CPA up 340%. Recommend pausing and shifting budget to the top two performers." That would have burned $3K by morning. The media buyer woke up to a problem already handled.
Your strategist reviews spend trends. Your account manager checks revenue attribution. Same Slack channel, same colleague, before anyone's first coffee.
Google Ads, Meta, Stripe. One message. No Looker, no Data Studio. Anomaly detection runs around the clock. Cross-platform reporting runs on autopilot.
5,700+ teams. SOC 2 certified. Your data never trains models.
"Viktor is now an integral team member, and after weeks of use we still feel we haven't uncovered the full potential." — Patrick O'Doherty, Director, Yarra Web
Quote of the week
I would say, more likely than not, 50% of all Super Bowl spots we watch this year will utilise generative AI in some facet."
Trending AI tools for marketers
Here are 4 trending tools to explore this week. Quick, practical upgrades to your AI marketing arsenal:
Flow: Hold a key, talk to any app, get text. Voice-to-prompt for daily marketing work.
Jet Admin: Build a customer-facing business AI agent in minutes.
Predis.ai: Turn product catalogues into social posts, captions and short videos.
Averi AI: Lead-finding and personalised outreach agent for small marketing teams.
TV, podcasts & streaming
How AI agents are transforming modern marketing strategy
An insightful conversation with Wharton's Stefano Puntoni on how chatbots and autonomous agents are reshaping search, discovery and brand strategy - watch it to it here.
AI training
How to launch a personalised ad creative in under 30 minutes using Hightouch
The overview:
Hightouch is an AI marketing agent that pulls from your brand assets to autogenerate on-brand ad creative without designer time.
Step-by-step:
Connect your sources
Link your Figma file, photo library and CDP through Hightouch's connectors panel. Authentication takes minutes if your stack is already in one place.Train the agent on your brand Upload one strong example of past creative so it learns your visual rules. Add three lines on tone, prohibited claims and your CTA style.
Define your audience Pick a customer segment from your warehouse using a natural-language prompt. Keep the segment narrow enough that the creative can speak directly to it.
Generate variants Ask Hightouch for five image variants and three captions for that segment. Reject anything off-brand before it goes any further down the workflow.
Approve and ship Send the winners straight to your ad platform of choice via the integration. Tag the launch so you can attribute lift back to this exact prompt.
Pro tip: Run the same prompt across three customer segments in parallel, then compare engagement before scaling. The model's "best" creative for one segment is almost never the best for another.
The weekly deep dive
Marketers are rebuilding their stack with AI-built apps, not SaaS
A wave of marketing teams is replacing licensed SaaS with internally generated AI tools, and it is reshaping where real martech value sits.
Churn is accelerating
Marketers are prototyping replacements with Cursor, Claude Code and similar agents in days, not quarters, sending churn climbing inside content, analytics and reporting tools. The teams making the switch are no longer waiting for vendor roadmaps and are shipping internal tools faster than the SaaS providers can patch them.
The cost line is changing shape
Spend is moving from per-seat SaaS contracts towards model usage and small engineering retainers, which makes budgets harder to forecast but cheaper at the margin. Finance teams are starting to ask CMOs to justify any seat-based renewal that an in-house prototype could replace.
Vendors are repositioning
Suite vendors are shifting from "the system of record" to "the connective tissue", betting on integrations and data gravity to keep marketers from going fully bespoke. Expect more aggressive bundling, free agent layers, and pricing changes designed to slow the bleed.
Our takeaway: If your team can prompt, your renewal cycle just got a lot more interesting; audit which SaaS tools could be replaced before procurement does it for you.
📖 Read the full article at MarTech
