Ad Age, the advertising industry's trade paper of record, published a reality check this week on the AI ad boom, asking the question every media planner hears from finance: where does the spend actually pay back?

The piece, headlined "AI advertising reality check: Where marketers are really seeing ROI," sits inside a genre the publisher keeps sharpening - cutting through glossy vendor releases to ask whether anyone has the receipts.

What is the headline really saying?

Three words carry the weight: "really seeing." That phrase is the tell. It implies a gap between vendor claims and what marketers actually measure on a P&L (profit-and-loss statement).

Return on investment - revenue retained after ad costs - is what every CMO (chief marketing officer) now has to defend. The article narrows in on marketers who can tie specific numbers to specific deployments, and discounts the rest.

The format is well-worn. Trade-press reality checks tend to spotlight practitioners with spreadsheet-level receipts and ignore the deck-driven claims. Ad Age has built an editorial lane on exactly this kind of audit.

Which AI use cases does the title favor?

AI in advertising splits into a working camp and a hype camp. The phrase "where marketers are really seeing ROI" telegraphs which buckets have moved past pilot status:

• Paid search copywriting. Generative AI (software that writes ad text from short prompts) producing headlines and descriptions at scale is the most mature ROI case, because search ads tie directly to clicks and conversions (completed purchases or sign-ups).

• Audience targeting and bid management. Automated real-time pricing for ad placement has worked these channels for years. AI rebranded existing math rather than inventing new lift.

• Product recommendation and on-site personalization. Recommending the right item to the right visitor reliably lifts average order value (the typical spend per checkout).

Which use cases does the title doubt?

Trade-press reality checks tend to land hardest on awareness-stage work - campaigns furthest from a clean click. Generative video, synthetic creator content (clips featuring AI-generated people), and AI-built brand assets rarely produce attribution models (the data trail crediting each touch with part of a sale) that survive a finance review.

Those campaigns can be visually striking. They usually cannot prove lift in a model an auditor will sign off on.

Why does this story land now?

CMO budgets carried AI spend into 2024 and 2025 planning cycles on the strength of vendor pitches about productivity and personalization. CFOs (the executives who control the purse) now want proof beyond the slide deck - attributed revenue, not engagement metrics (vanity numbers like impressions or time-on-page that don't tie to dollars).

The framing matters. A skeptical Ad Age piece can quietly deflate an entire product category's growth narrative. The readership - agency leaders, brand marketers, platform sellers - reads this kind of story closely, and quietly rebalances spend toward whatever the article signals is safe.

The bottom line

For anyone walking into a quarterly review with an AI advertising proposal: bring the math, not the manifesto. The headline itself, "reality check," is the lesson.

Ad Age ran against the prevailing optimism of the AI ad moment. It asked for less magic, more auditable returns. That trade-press framing is worth reading twice before the next budget cycle closes.