AI-generated ads get attention, but is it the good kind?

Canadian brands and agencies are using generative AI selectively, mainly in pre-production, minor visual edits, and low-profile social content, due to consumer backlash and IP risks. Studies cited in the article suggest disclosed AI ads can reduce effectiveness, with one NYU/Emory analysis finding click-through rates fell as much as 31.5% when AI use was disclosed. The piece also highlights emerging AI production studios and a likely shift toward lower-cost, AI-assisted ad creation, though most finished work still uses traditional production crews.

Analysis

The economic value of generative AI in advertising is not in replacing full creative pipelines overnight; it is in collapsing pre-production and versioning costs first. That creates an immediate winner set among workflow vendors and production-enablement platforms, while the losers are the labor-intensive middle layers: small studios, photo/retouch shops, and agencies whose margin models depend on billable hours. The second-order effect is that AI adoption will likely show up first as gross-margin expansion at agencies and SaaS vendors before it becomes visible in top-line growth, because clients will push for lower prices while demanding the same output quality.

For ADBE and RNG, this is more subtle than a simple “AI tailwind” trade. Adobe benefits if creative teams standardize on its suite as the default orchestration layer for AI-assisted iteration, but there is a real risk that commoditized AI generation reduces seat expansion and shifts value away from software toward independent model providers. RingCentral is a cleaner beneficiary of the broader corporate reallocation toward AI-generated content and AI-enabled customer experience; the market may be underappreciating that firms using AI to compress content cycles usually also increase spend on cloud collaboration, review, and approval workflows.

The biggest near-term risk is reputational, not technological: disclosure can mechanically reduce ad efficacy, but non-disclosure raises legal and brand-safety risk, especially around IP and likeness. That means the adoption curve should be choppy over the next 3-12 months, with headline risk around any high-profile consumer backlash or infringement claim. Contrarian take: the market may be overestimating how fast fully AI-generated ads scale in consumer brands, but underestimating how fast AI becomes embedded in invisible upstream production, where the ROI is easier to defend and less exposed to consumer scrutiny.

Key Decisions for Investors

• Long ADBE vs WPP, 3-6 month horizon: buy ADBE on pullbacks and short WPP as a relative-value pair. Thesis is that software captures workflow standardization while agencies absorb pricing pressure; target 10-15% spread with stop if WPP shows accelerating new-business wins.
• Initiate a small long RNG position or call spread into the next 1-2 quarters. AI-driven content ops and collaboration should lift enterprise workflow intensity; asymmetric upside if management frames AI as a seat-retention tool rather than a pure communications product.
• Buy KO put spreads or maintain a tactical underweight for 1-3 months around any AI-campaign controversy. Brand-sensitive consumer staples are vulnerable to reputational overhang even when the operational impact is negligible; risk/reward favors hedging headline risk rather than calling for a fundamental earnings hit.
• Long V on dips, 6-12 month horizon. AI-generated campaign personalization should increase transaction-level marketing efficiency and support merchant activity; this is a quieter beneficiary with limited disclosure risk and better multiple durability.
• Avoid chasing pure-play AI creative names until IP litigation clarity improves. The first meaningful catalyst is likely a legal/brand backlash event, not a revenue beat; use any selloff in ADBE or V as entry, but fade speculative moves in smaller studios.