---
id: "question-liability-third-party-agents"
type: "open-question"
source_timestamps: ["§ 4. Observe how your brand shows up in agent ecosystems."]
tags: ["legal", "liability", "hallucinations"]
related: ["claim-brand-failure-not-system-error"]
resolution_path: "Future regulatory rulings or high-profile lawsuits establishing precedent for AI agent intermediary liability."
source_url: "https://hbr.org/2026/02/how-brands-can-adapt-when-ai-agents-do-the-shopping"
source_title: "How Brands Can Adapt When AI Agents Do the Shopping"
sources: ["geo"]
sourceVaultSlug: "hbr-seg-geo"
originDay: 3
articleStem: "hbr-ext-14-brands-adapt-ai-shopping"
sourceUrl: "https://hbr.org/2026/02/how-brands-can-adapt-when-ai-agents-do-the-shopping"
sourceTitle: "How Brands Can Adapt When AI Agents Do the Shopping"
---
# Who Holds Legal Liability for Third-Party Agent Hallucinations?

**Open question.** While the article argues consumers *perceive* platform hallucinations as brand failures ([[claim-brand-failure-not-system-error]]), it leaves open the question of **legal and financial liability**. If a general-purpose AI (like [[entity-chatgpt-d14]]) promises a customer a **non-existent discount** on a brand's product, is the **brand legally obligated to honor it**, or is the **AI platform** liable?

**Resolution path.** Future **regulatory rulings** or **high-profile lawsuits** establishing precedent for AI-agent intermediary liability.

> **Enrichment / counter-perspective.** Legal scholars and regulators are actively debating whether platforms, brands, or intermediaries bear liability for AI-generated misrepresentations. Some argue that **platforms providing the AI should carry primary responsibility**, especially where brands have limited control over how they are described — which directly challenges the article's assumption that misrepresentation is always *treated* as a brand failure. Relevant governance context: OECD AI Principles, the EU AI Act, and the NIST AI Risk Management Framework.


## Related across articles
- [[question-ai-liability-governance]]
- [[concept-transaction-grade-governance]]
- [[claim-brand-failure-not-system-error]]
