📚 New: 8 free AI marketing playbooks inside

Welcome to this week's AI + Marketing Weekly! New section inside: Builds & Playbooks. Plus: GPT-5.6 dropped, Google labels AI-made ads, and Fable 5 extended.

Carley Lake

Jul 14, 2026

Fable’s back…back again. NurPhoto via Getty Images, via Forbes.

~The AI + Marketing Weekly~

what you need to know in ai + marketing (15-second version):

OpenAI launched GPT-5.6 in three tiers. Sol (flagship), Terra (mid-tier, half the cost of GPT-5.5), Luna (cheap everyday). Sol is 54% more token-efficient for coding. ChatGPT Work, a new agent that handles multi-hour projects across apps, launched alongside it. (TechCrunch)
Google will now disclose which ads are made with AI. A new “How this ad was made” panel is rolling out across Search, YouTube, and Discover. Ads built with Google’s AI tools get labeled automatically. Third-party AI tools require manual disclosure. (TechCrunch)
SEO teams risk losing GEO ownership to PR and content. Authority, reputation, and third-party validation now drive AI answer inclusion more than keywords. The skill set for generative engine optimization looks more like PR than traditional SEO. (MarTech)
The first AI commerce rankings are live. Digital Commerce 360 and ReFiBuy launched the first industry ranking of retailers by AI channel performance, measuring share-of-recommendation across shopping agents and answer engines. (MarTech)
BCG says the AI leader-laggard gap is widening. Top-quartile adopters invest more in operating model redesign. Followers over-invest in tools and under-invest in architecture. The gap is growing, not closing. (MarTech)
Anthropic extended Claude Fable 5 access to July 19. Pro, Max, Team, and Enterprise subscribers keep Fable 5 at up to 50% of weekly usage limits, with a rate limit boost. After July 19, it shifts to pay-per-use credits ($10/$50 per million tokens). Anthropic says it aims to restore Fable 5 as a standard subscription feature once capacity allows. (Forbes)
New this week: Builds & Playbooks. You asked for more “show me how to actually do it” content, so I’m adding a new section! This week: 8 free guides, templates, and walkthroughs for marketers building with AI, from Claude Code tutorials to a 44-skill marketing plugin. Scroll down to check it out. Thanks for the suggestions, and keep them coming!

🗞️ top ai + marketing news

openai drops a three-tier model family

OpenAI launches GPT-5.6 with Sol, Terra, and Luna
TechCrunch | ~5-min read

GPT-5.6 launched July 9 after a White House cybersecurity delay. Three tiers: Sol (flagship, best coding model), Terra (competitive with GPT-5.5 at half the cost), Luna (cheap everyday tasks). Sol is 54% more token-efficient. Alongside it, ChatGPT Work launched as an agent that manages multi-hour projects across desktop apps, files, and web.

Why it matters: OpenAI just gave marketing teams a price ladder…Luna for daily grunt work, Terra for campaigns, Sol for the hard stuff. The teams that route tasks to the right tier will spend a fraction of what teams defaulting to the top model spend.

google labels ai-made ads

Google will now disclose which ads are made with AI
TechCrunch | ~3-min read

Google is expanding AI ad disclosure beyond election ads to all ad formats. A new “How this ad was made” panel shows up in My Ad Center across Search, YouTube, and Discover. Ads created with Google’s generative tools get labeled automatically. Third-party AI tools require manual disclosure from advertisers.

Why it matters: Consumers already say they trust brands less when they know AI made the ad and now Google is telling them. If your creative is AI-generated, plan for that label being visible.

trust is now the top AI search signal

Why AI search makes trust your most important visibility signal
MarTech | ~4-min read

AI models rely on authoritative, credible sources when generating answers. With 68% of searches ending in no clicks, website traffic alone no longer equals success. Brands need to optimize across earned media, reviews, and third-party references, not just owned properties. The skill set looks more like PR than traditional SEO.

Why it matters: If your SEO team owns AI search visibility, they need to start thinking like PR.

OpenAI set to expand ads to France, Germany and Ireland
Digiday | ~4-min read

OpenAI is hiring regional leadership in Paris, Munich, and Dublin to bring ChatGPT ads to 10 markets globally, less than a year after launching the ad pilot. Criteo says it doubled advertiser count on the OpenAI pilot in a single month. Over 2,000 brands now run ads through Criteo’s ChatGPT integration.

Why it matters: ChatGPT ads went from zero to 10 countries in under a year. If your brand sells internationally, this is no longer a US-only experiment.

the best campaigns of 2026 sold honesty

The best brand campaigns of H1 2026 sold honesty. Are consumers buying?
Marketing Dive | ~5-min read

Coca-Cola, Burger King, Unilever, and others doubled down on authenticity in the first half of 2026, copping to past mistakes and pushing back against the rise of AI. Anthropic’s Claude “no ads” Super Bowl campaign won the Super Clio. The question: can sincerity be a competitive advantage when AI is commoditizing everything else?

Why it matters: As AI floods the market with content, the brands that stand out are the ones willing to be human about it. Honesty is becoming a positioning strategy and not just a value.

the first ai commerce scoreboard is live

Digital Commerce 360 and ReFiBuy launch the first AI commerce rankings
ChannelX | ~4-min read

The first ranking of Top 1000 retailers by AI readiness is live. Each retailer gets a 0-100 score across four signals: bot friendliness, AI source traffic volume, diversity of AI sources, and 90-day momentum. The surprise: the retailer ranked #814 by online sales is #1 in AI readiness. Early leaders include Online Labels, Everlane, and Brooklinen.

Why it matters: Scale alone does not equal AI readiness. Small brands with clean catalogs and bot-friendly sites are outranking giants.

the ai gap is getting worse, not better

BCG: how AI leaders create competitive advantage
BCG | ~5-min read

BCG’s latest research shows top-quartile AI adopters invest more in operating model redesign than tools. Followers do the opposite: they buy tools and skip the architecture. The leader-laggard gap is widening, not closing.

Why it matters: Buying AI tools without redesigning how your team works is like buying a race car and driving it in a parking lot. The gap is in the operating model and not the tech.

📚 builds & playbooks

New section! A lot of you have been asking for more hands-on, “show me how to actually do it” content. So I’m adding Builds & Playbooks: the best templates, 101 guides, and walkthroughs I find each week for marketers building with AI. Enjoy! 😊

templates & starter kits

Emily Kramer (MKT1) - “Build a marketing strategy skill in Claude Code”
Walk through seven exercises (ICP prioritization, positioning, revenue levers, Big Bet campaigns) and embed your strategy directly into Claude as a reusable skill. Every brief and campaign review references it automatically.

Sabrina Ramonov - “Set up your AI marketing team in Claude (7 free skills)”
Seven free skills (brand-brief, post-writer, post-grader, post-scheduler, viral-hooks, repurpose, content-coach) that handle the full content pipeline from ideation through cross-platform publishing.

Claude Directory - 44-skill marketing plugin
Forty-four specialist skills across eight pods: Foundation, Content, SEO/AEO, CRO, Channels, Growth, Intelligence, and Specialized. Install the whole directory and route to the right skill per task.

101s

Robert Gillespie - “The marketer’s guide to Claude Code (non-techie version)”
Reframes Claude Code for non-technical marketers. Real example: a growth marketer built marketing automation without coding experience. Covers the three access methods (Desktop, Terminal, VS Code) and practical use cases like automated ad variants and SEO audits.

walkthroughs

Sabrina Ramonov - “The ultimate Claude Code tutorial”
A 90-minute course that builds a personalized AI social media manager step by step. Covers skills, hooks, subagents, permissions, and a /plan-week skill that generates week-long content calendars with custom visuals.

Stack & Scale - “How I turn Claude Code into a full marketing team”
Three workflows: competitive intelligence briefs, content repurposing engine (one blog post into multi-channel assets), and automated weekly executive dashboards. Treats Claude like a contractor, not a chatbot.

Ryan Doser - “How to use Claude Code for marketing”
Beginner setup guide covering folder structure, skills files for persistent brand voice, and thumbnail generation using Claude Code with Google’s image API through OpenRouter.

👩‍💻 thought-leader highlights

Phyllis Fang - “6 ways to turn your customers into your champions”

Phyllis laid out a progression from logo walls to full customer evangelism, using Michael Scott as the case study (stay with her, it works). The real insight: most companies stop at “Can we use your logo?” but real proof is showing buyers who trusts you, why they trust you, and how they talk about you when you’re not in the room.

Axelle Malek - “You’re still guessing which Claude model to use”

Axelle published a 40-second routing test: Haiku for quick chat, Sonnet for everyday tasks, Opus for deep work, Fable for your hardest 10%. Her rule: use Fable for one or two strategic turns, then switch back to Opus. Long chats get expensive because Claude re-reads the whole thread every turn.

Monica Abrams - “Here are the cases where you should NOT use AI”

Monica pushed back on the “AI everything” wave with a list of when AI should take a backseat. After a year of building a business with AI daily (and failing with it along the way), she mapped where it helps and where it hurts. The comments turned into a crowdsourced list of AI boundaries.

Nate Herk - “I tested GPT 5.6 Sol vs Fable 5. Here’s what you need to know.”

Nate ran a head-to-head: Sol won 24-3 on quick API tasks and cost $16 versus $63 for the same work. But Fable is “the manager” for strategic and creative work. His framework: if the task is a 5, you don’t need a model that’s a 10. Match the model to the job and your unit economics change overnight.

Nicole Leffer - “4 things marketers need to know about ChatGPT Work”

Nicole trained a marketing team on ChatGPT Work and shared the key takeaways. The biggest one: pre-plan in a regular chat before you go to Work. Your Work allowance is limited, so don’t waste it figuring out what you want. Flush out the ask, define success criteria, and have Chat help you write the prompt. Good Work begins before you put the prompt in.

🛠️ latest ai + marketing tools

ChatGPT Work
AI agent that manages multi-hour projects across desktop apps, files, and web.
Launched July 9 alongside GPT-5.6. Creates slides, sheets, and docs from goals rather than prompts. Includes enterprise security and governance controls.

Guideline Ad Intelligence for AI Platforms
Verified ad spend tracking across ChatGPT and Perplexity.
Launched July 9. Captures transaction-level advertising activity on AI platforms. Covers $200 billion in annual media investment across 65 countries.

ActiveCampaign Google Ads Connector
Links first-party customer data directly to Google Ads campaigns.
Launched July 10. Uses machine learning to match customer data to Google Ads profiles and build campaign assets through automated prompts.

Profound Ads Studio + AIM Platform
AI search advertising and visibility management platform.
Launched July 9. Extends the Profound Index (daily AI search visibility benchmark) with ad buying and optimization tools for AI answer engines.

Claude Fable 5 (extended access)
Anthropic’s most capable model, now available to subscribers through July 19.
Pro, Max, Team, and Enterprise plans get Fable 5 at up to 50% of weekly usage limits with a rate limit boost. After July 19, shifts to pay-per-use. Anthropic aims to make it a standard subscription feature once capacity allows.

Claude Reflect
Dashboard that visualizes how you interact with Claude across sessions.
Launched July 9. Shows topics discussed, usage patterns, and task categories. Helps you see where you’re actually getting value and where you’re underusing the tool.

Stensul with Brand Governance
Content creation platform with AI brand compliance filters.
Launched July 9. Enforces brand guidelines across AI-generated marketing content. Designed for teams scaling content production while maintaining consistency.

💼 cool ai + marketing jobs

monday.com - Agentic Platform Marketing Lead, CRM

NVIDIA AI - Director, Strategic Marketing Operations

OpenAI - Growth Marketing, Creative System and Optimization

Google - Principal Lead, Go-To-Market AI Tools

Microsoft - Senior Director - Integrated and Campaign Product Marketing, AI Business Solutions

Carley Lake

Discussion about this post

Ready for more?