Google is quietly changing the operating system for a large share of paid search. During September 2026, campaigns using campaign-level broad match or legacy Automatically Created Assets are being migrated in place to AI Max for Search campaigns. The shift matters because it changes who controls query expansion, ad copy, and—eventually—landing-page selection: the advertiser, or Google’s models.

This is not a new campaign launch that marketers can ignore until the next planning cycle. Google stopped allowing new legacy configurations on August 3, and its September migration is rolling out campaign by campaign. Google says the affected campaigns will keep their IDs and equivalent settings, but the defaults now make AI-mediated matching a permanent part of the account. The right response is not panic or blind acceptance; it is measurement before automation removes the choice.

What Google AI Max is changing now

Google’s current migration has two cohorts. Campaigns using Automatically Created Assets receive AI Max with search term matching and text customization turned on by default. Campaigns using the campaign-level broad match setting receive search term matching by default. Final URL expansion is not automatically enabled for either cohort, while existing brand inclusions and exclusions carry over.

Dynamic Search Ads are on a later track. Google’s updated timeline moves their automatic migration to February 2027, giving teams a short but useful window to test the new structure before legacy DSA creation disappears. Google’s own recommendation is direct: “We recommend transitioning to AI Max now to maintain full control over your campaign setup” (PPC Land’s report on the Google Ads notice).

The upside is real—but it is a modeled promise

Google says advertisers using the full AI Max feature set—search term matching, text customization, and final URL expansion—see an average 7% increase in conversions or conversion value at a similar CPA or ROAS compared with search term matching alone (Google Ads & Commerce Blog). Google also says hundreds of thousands of advertisers are already scaling Search campaigns with AI Max.

That claim is useful as a hypothesis, not a forecast for every account. The official post does not publish sample size, test duration, geography, confidence intervals, or a breakdown by industry. AI Max can discover incremental intent and generate more relevant variants, but the same expansion can spend into queries that look plausible while carrying weak commercial intent. Treat the 7% figure as a reason to run a controlled experiment—not permission to remove guardrails.

Why traffic quality is the CEO-level risk

Automation changes the economics of wasted spend as well as the economics of growth. In a study of more than 414 million clicks collected from October 2025 through June 2026, ad-quality firm Lunio reported that retail Google Search campaigns with AI Max enabled had a 5.28% invalid-traffic rate in Q2 2026, versus 3.07% for standard Search campaigns—a 72% difference (PPC Land’s summary of the Lunio report).

The finding is not proof that AI Max caused the gap. It is a warning that conversion volume alone is not a sufficient success metric when matching expands. Lunio defines invalid traffic broadly, including bots, automated activity, competitor clicks, and accidental clicks. Finance leaders should ask for qualified-lead rate, downstream revenue, search-term quality, and invalid-traffic controls alongside CPA or ROAS.

A 30-day AI Max readiness plan

Businesses with affected campaigns should take four steps:

  1. Inventory the change. Export campaign settings, search terms, asset combinations, landing pages, negative keywords, brand controls, and conversion actions before and after migration. Google’s reporting now labels relevant traffic with the match type “AI Max” and provides a search-terms-and-ad-combinations view.
  2. Protect the boundaries. Review URL inclusions and exclusions, brand lists, location settings, text guidelines, and pinned assets. Keep sensitive products, regulated claims, and low-margin categories out of automated expansion until they pass human review.
  3. Run a clean test. Use Google’s experiment tools where available. Hold budget, offer, landing-page quality, and conversion definitions steady; compare qualified pipeline and revenue, not just clicks or platform-reported conversions.
  4. Fix the feedback loop. Import offline conversions and revenue values, separate branded from nonbranded demand, and review search terms at least weekly during the transition. If the account cannot tell Google which leads become customers, the model is optimizing a proxy.

AI Max is a turning point because paid search is moving from keyword management to intent management. The winners will not be the teams that surrender control fastest; they will be the teams that give automation better business signals and better constraints.

Real Internet Sales helps businesses adapt paid search, AI visibility, and conversion systems to this new operating model. Call 803-708-5514 or visit realinternetsales.com to build an AI-ready marketing plan before platform defaults make the decisions for you.