Museum Wire
Law 0 · Katie's LawEvery system is shaped by the human drive to do less work. This is not a flaw. It is the economic force that produces all software — and all software failure.◆Law I · Boundary CollapseWhen data crosses into a system that interprets structure, without being constrained, it becomes executable.◆◆ 2026 IncidentAxios. 70 Million Downloads a Week. North Korea Inside.◆Law II · Ambient AuthorityWhen a system trusts the presence of a credential instead of verifying the intent behind it, authentication becomes indistinguishable from authorization.◆◈ AXM-001Set Theory — Membership, Boundaries, and Belonging◆Law III · Transitive TrustWhen a system inherits trust from a source it did not verify, the attack surface extends to everything that source touches.◆◆ 2026 IncidentClaude Code — The Accept-Data-Loss Flag◆Law IV · Complexity AccretionSystems do not become complex. They accumulate complexity — one reasonable decision at a time — until no single person can hold the whole in their head.◆Law V · Temporal CouplingCode that assumes sequential execution, stable state, or consistent timing will fail the moment concurrency, scale, or latency proves the assumption wrong.◆◆ 2026 IncidentCopy Fail — 732 Bytes to Root on Every Linux Distribution◆◈ AXM-002Boolean & Propositional Logic — True, False, and the Excluded Middle◆Law VI · Observer InterferenceWhen the system that monitors health becomes a participant in the system it monitors, observation becomes a failure vector.◆◆ 2025Amazon Kiro — The 13-Hour Outage◆◆ 2025Operation Chrysalis: The Notepad++ Supply Chain Hijack◆◆ 2025Replit Agent — The Vibe Code Wipe◆◆ 2025Shai-Hulud — The npm Worm That Ate Its Own Ecosystem◆◆ 2024Air Canada Chatbot — The Policy That Wasn't◆◆ 2024Change Healthcare — One-Third of US Healthcare, One Missing MFA◆◆ 2024CrowdStrike — The Security Update That Broke the World◆◆ 2024Google Gemini Image Generation — The Six-Day Pause◆◆ 2024XZ Utils — The Two-Year Infiltration◆◆ 20233CX — The Supply Chain That Ate Another Supply Chain◆◆ 2023Amazon Prime Video — The Per-Frame State Machine◆◆ 2023Bing Sydney — The Chatbot That Went Rogue◆◆ 2023Samsung ChatGPT Leak — The Employee Who Pasted the Secret◆EFFODE · LEGE · INTELLEGE◆Law 0 · Katie's LawEvery system is shaped by the human drive to do less work. This is not a flaw. It is the economic force that produces all software — and all software failure.◆Law I · Boundary CollapseWhen data crosses into a system that interprets structure, without being constrained, it becomes executable.◆◆ 2026 IncidentAxios. 70 Million Downloads a Week. North Korea Inside.◆Law II · Ambient AuthorityWhen a system trusts the presence of a credential instead of verifying the intent behind it, authentication becomes indistinguishable from authorization.◆◈ AXM-001Set Theory — Membership, Boundaries, and Belonging◆Law III · Transitive TrustWhen a system inherits trust from a source it did not verify, the attack surface extends to everything that source touches.◆◆ 2026 IncidentClaude Code — The Accept-Data-Loss Flag◆Law IV · Complexity AccretionSystems do not become complex. They accumulate complexity — one reasonable decision at a time — until no single person can hold the whole in their head.◆Law V · Temporal CouplingCode that assumes sequential execution, stable state, or consistent timing will fail the moment concurrency, scale, or latency proves the assumption wrong.◆◆ 2026 IncidentCopy Fail — 732 Bytes to Root on Every Linux Distribution◆◈ AXM-002Boolean & Propositional Logic — True, False, and the Excluded Middle◆Law VI · Observer InterferenceWhen the system that monitors health becomes a participant in the system it monitors, observation becomes a failure vector.◆◆ 2025Amazon Kiro — The 13-Hour Outage◆◆ 2025Operation Chrysalis: The Notepad++ Supply Chain Hijack◆◆ 2025Replit Agent — The Vibe Code Wipe◆◆ 2025Shai-Hulud — The npm Worm That Ate Its Own Ecosystem◆◆ 2024Air Canada Chatbot — The Policy That Wasn't◆◆ 2024Change Healthcare — One-Third of US Healthcare, One Missing MFA◆◆ 2024CrowdStrike — The Security Update That Broke the World◆◆ 2024Google Gemini Image Generation — The Six-Day Pause◆◆ 2024XZ Utils — The Two-Year Infiltration◆◆ 20233CX — The Supply Chain That Ate Another Supply Chain◆◆ 2023Amazon Prime Video — The Per-Frame State Machine◆◆ 2023Bing Sydney — The Chatbot That Went Rogue◆◆ 2023Samsung ChatGPT Leak — The Employee Who Pasted the Secret◆EFFODE · LEGE · INTELLEGE◆
Keyboard Navigation
W↑
A←
S↓
D→
or arrow keys · M for map · Q to exit
← Back to Incident Room
2024legal liabilityconsumer

Air Canada Chatbot — The Policy That Wasn't

Air Canada's customer service chatbot told a grieving customer he could purchase a full-price bereavement ticket and apply for a discount retroactively — a policy that did not exist. Air Canada argued the chatbot was a separate legal entity responsible for its own statements. The Canadian Civil Resolution Tribunal ruled against Air Canada, holding the airline liable for its chatbot's incorrect policy representation.

3 min read
Root Cause

The chatbot generated a plausible-sounding refund policy that contradicted the airline's actual policy. Air Canada deployed the chatbot as a customer-facing policy authority without implementing guardrails to prevent policy confabulation.

Aftermath

First major court ruling establishing that a company is liable for its AI chatbot's statements. Accelerated enterprise AI governance, disclaimer requirements, and the question of chatbot authority scope in customer-facing deployments.

The Incident

In November 2022, Jake Moffatt's grandmother died. He needed to fly from Vancouver to Toronto for the funeral. He consulted Air Canada's customer service chatbot before booking.

The chatbot told him that Air Canada offered bereavement fares, and that he could purchase a full-price ticket, travel, and then apply for the bereavement discount retroactively within 90 days.

He did exactly that. He submitted the retroactive refund request. Air Canada denied it — because the policy the chatbot described did not exist. Air Canada's actual bereavement policy required the discounted fare to be applied at the time of booking, not after the fact.

Moffatt took Air Canada to the Canadian Civil Resolution Tribunal. Air Canada argued that its chatbot was effectively a separate entity — that the airline was not responsible for information the chatbot provided independently — and that Moffatt should have checked the official website.

The Tribunal ruled against Air Canada.

The Legal Finding

The Tribunal's reasoning: Air Canada deployed a chatbot as a customer service interface. Customers interacting with that interface are entitled to expect that information it provides reflects the company's actual policies. The chatbot is not a separate entity — it is Air Canada's representative. Air Canada is liable for what its representative says.

This was the first major court ruling establishing that a company is liable for its AI chatbot's incorrect statements, regardless of whether those statements were generated autonomously by the model.

The Pattern

This incident is Ambient Authority (Law II) at the deployment layer: the chatbot was granted customer-facing authority to speak about Air Canada's policies. No constraint was placed on what policies it could describe. The chatbot generated a plausible-sounding policy and stated it as fact. The authority to speak on behalf of Air Canada was real. The policy was fabricated.

This connects to [The Confident Confabulator](/exhibits/the-confident-confabulator): the chatbot did not know that the retroactive policy didn't exist. It generated the most plausible-sounding description of a bereavement travel policy, given its training. The plausibility of the output was not correlated with its accuracy.

The Archaeologist's Note

Air Canada's legal argument — "the chatbot is a separate entity, not our responsibility" — is the single most important statement in the museum's AI section. It is the moment a corporation deployed an AI customer service agent, allowed it to state policy as fact, charged a customer based on that policy, denied the claim, and then argued in court that the agent's statements were not the company's statements.

The tribunal's rejection of that argument is the beginning of AI liability jurisprudence. Every enterprise deploying a customer-facing AI agent now operates in the shadow of this ruling.

Katie's Law: The shortcut was deploying a chatbot that could say anything about your policies without verifying that what it said matched your policies. Because implementing that verification was hard. The court ruled that the shortcut cost exactly what it should have.
Techniques
ambient authoritypolicy hallucinationcorporate chatbot liability