---
id: "contrarian-advanced-ai-rationality"
type: "contrarian-insight"
source_timestamps: ["\\\"§ Expect more advanced models to be skeptical of marketing tactics", "not indifferent to them.\\\""]
tags: ["ai-behavior", "marketing-theory", "contrarian"]
related: ["concept-algorithmic-skepticism", "quote-persuasion-penalty"]
challenges: "The conventional view that advanced AI will act as a perfectly rational, indifferent utility maximizer in commercial settings."
sources: ["geo"]
sourceVaultSlug: "hbr-seg-geo"
originDay: 3
articleStem: "hbr-tier2-06-ai-shopping-agents"
sourceUrl: "https://hbr.org/2026/05/research-traditional-marketing-doesnt-work-on-ai-shopping-agents"
sourceTitle: "Research: Traditional Marketing Doesn’t Work on AI Shopping Agents"
---
# Advanced AI doesn't just ignore marketing; it penalizes it

**Contrarian insight — challenges:** the conventional view that advanced AI will act as a perfectly rational, *indifferent* utility maximizer in commercial settings.

**Conventional wisdom:** As AI models become more capable, they will become perfectly informed "utility maximizers" that simply **ignore** irrelevant marketing noise, focusing purely on value.

**What the research shows:** The opposite. Advanced models **actively penalize** overt persuasion cues, interpreting them as signals of low quality or manipulation. More persuasion can produce *fewer* selections (see [[concept-algorithmic-skepticism]] and [[quote-persuasion-penalty]]). Indifference is not the destination — **skepticism** is.

**Enrichment / calibration:** Independent research (ACES/ACE) confirms that some promotional overlays backfire and that presentation biases differ in direction and magnitude between model generations. But the *strong* form — that advanced models **systematically** treat overt persuasion as a negative quality signal — is best held as a **hypothesis supported by initial data**, not a settled law. The safer statement: advanced models develop *systematic biases and skepticism patterns*, which is emergent behavior, not guaranteed rationality.

**Related:** [[concept-algorithmic-skepticism]] · [[quote-persuasion-penalty]] · [[concept-reasoning-vs-non-reasoning-models]]


## Related across articles
- [[contrarian-bot-rationality]]
- [[contrarian-sentiment-optimization]]
- [[claim-sponsored-penalty]]