“AI marketing” has become the most overloaded phrase in our industry. A new explainer from PPC Land cuts through it with a definition worth stealing: it is marketing where a statistical model, not a person, makes the decision.
You hand over an objective, a budget and some constraints. Software fills in the specifics. That is the whole idea — and it is why adoption stats swing wildly depending on who is asking.
Three layers, not one thing
PPC Land splits AI marketing into three technical generations that now run side by side.
- Prediction: the oldest layer, living inside the auction. Google groups four strategies under Smart Bidding and describes more than 20 signals read at auction time — device, OS, location, time of day, browser, query. Meta’s Advantage+ suite covers audience, placement, budget and creative from an objective and a country.
- Generation: asset production. Google finished the generative rollout of automatically created assets in February 2024 for US and UK English advertisers, refreshing assets at least every 48 hours and serving machine copy only when it is predicted to beat yours. Asset Studio launched 10 September 2025. Meta added branded text and image generation at Cannes Lions on 17 June 2025.
- Agents: software that plans and executes. The Ad Context Protocol arrived in October 2025 from a supply-side coalition including Scope3, Yahoo and PubMatic. IAB Tech Lab named its rival umbrella AAMP on 26 February 2026.
The manual exits are closing
This is the part that should sharpen your planning. Platforms are not just offering automation, they are withdrawing the alternative.
Microsoft retires max CPC fields on new campaigns using automated strategies from 1 October 2026. OpenAI made an automated strategy the default in new ChatGPT Ads ad groups in mid-August 2026. Google started converting campaigns running automatically created assets or campaign-level broad match into AI Max for Search on 1 September 2026.
If your team’s edge is manual bid tuning, that edge has an expiry date.
Adoption is high, returns are not
IAB Europe’s pan-European survey (18 September 2025, roughly 95 responses) found 85% of companies using AI marketing tools — 64% for targeting, 61% for content generation. Ad tech firms reported the best results, with 60% citing KPI improvements versus 48% for agencies.
Then the reality check. TransUnion found only 53% of marketers report meaningful ROI, with just 36% rating their data and process readiness as high. Typeface research in June 2026 found enterprise campaign timelines getting longer as adoption rose. StackAdapt, published 19 August 2026, recorded 91% tool usage but only 6% of marketers acting on in-platform AI recommendations — a gap the research pins on unresolved accountability, not capability.
Accuracy is shaky too. WordStream research from 10 July 2025 found 20% of AI answers to PPC questions contained inaccurate information, with Google AI Overviews at 26% wrong and Gemini at 6%.
Money is moving anyway
Mediaocean’s November 2025 survey of 320 professionals found 54% planning to increase AI media investment against 47% for search — the first time a nascent channel overtook search in that series. EMARKETER, in a 4 June 2026 report by Nate Elliott, projects US AI advertising at $32.03 billion in 2026 rising to $68.25 billion by 2030, with more than 80% of it sitting beside AI-generated content rather than inside chatbots.
Vendor forecasts diverge by an order of magnitude. PubMatic’s Rajeev Goel expects 25% of digital advertising to execute autonomously by 2028. Magnite’s Michael Barrett caps 2027 protocol-based agentic spend near $700 million. Neither has independent corroboration.
Disclosure has a price tag
Article 50 of the EU AI Act became applicable on 2 August 2026. Systems already on the market have until 2 December 2026 for the machine-readable marking obligation. Google has pushed AI ad labelling responsibility entirely onto advertisers.
And labels cost you: IAB’s January 2026 transparency framework cites NYU research showing AI labels cutting click-through rates by 31.5%, alongside a 37-point perception gap between advertisers and consumers.
What to do this quarter
Audit which of the three layers you actually use, and be honest about it. Fix data readiness before buying more tools — that 36% readiness figure explains most of the ROI gap. Build audit logs and spend caps before agents touch live budgets; agencies are already metering agents by the token after one connected TV test returned five times a $25,000 spend. And map your EU disclosure triggers now, not in November.
Source: PPC Land
