INSIGHTS

Decision Visibility
Case Studies

Real-world cases where organisations were challenged to explain consequential decisions.

Each case highlights a question Irish boards should be asking: Could we explain this if it happened here?

CASE 001WORKDAY
Read Case Study

Workday Recruitment Decisions — The Decision Visibility Gap in AI Act Compliance Ireland

In Mobley v. Workday, Inc. (N.D. Cal.), a job applicant alleged that Workday's AI-powered screening tools contributed to discriminatory hiring outcomes. The case was allowed to proceed, with the court considering whether Workday could bear responsibility as an agent acting on behalf of employers. Subsequent developments included collective-action certification around age-discrimination claims. The case remains ongoing.

DIGITALomnibus LENS: This is the Decision Visibility Gap — the difference between understanding individual systems and being able to explain the decisions that emerge between them. "One Decision. Multiple Systems."

CASE 002AIR CANADA
Read Case Study

Air Canada Chatbot — Who Owns the Outcome of an AI-Assisted Decision?

In Moffatt v. Air Canada (2024 BCCRT 149), a customer asked the airline's AI chatbot about bereavement fares. The chatbot provided incorrect information about the refund policy. When the customer sought to rely on that information, Air Canada argued the chatbot was a separate entity responsible for its own statements.

DIGITALomnibus LENS: The issue was not the chatbot itself, but the lack of clear Decision Visibility across the systems and processes that generated the final response.

CASE 003DUTCH GOVERNMENT
Read Case Study

Dutch Child Benefits Scandal — What Happens When Nobody Can Reconstruct the Decision?

The Dutch government deployed an automated risk assessment system to detect benefit fraud. The system incorrectly flagged thousands of families, disproportionately affecting minority groups, leading to devastating financial consequences.

DIGITALomnibus LENS: At its core, this was a failure of Decision Visibility across interconnected systems.

CASE 004AMAZON
Read Case Study

Amazon AI Recruiting Tool — When Automated Decisions Lack Visibility

Amazon developed an AI-powered recruiting tool intended to streamline hiring by reviewing resumes and scoring candidates. The system was trained on historical data from the company's previous hiring patterns.

DIGITALomnibus LENS: The core issue was not malice, but a complete lack of Decision Visibility. The system operated as a black box with no clear way to reconstruct or explain individual outcomes.

CASE 005PIZZA HUT / DRAGONTAIL
Read Case Study

Pizza Hut Dragontail AI — When Operational Decisions Go Wrong

Chaac Pizza Northeast, a major Pizza Hut franchisee operating over 110 locations, alleges that it was required to adopt Dragontail (now Yum! Connect), an AI-powered delivery management system designed to optimise order dispatch and timing. According to reporting by Business Insider, the franchisee claims the system caused significant operational disruption.

DIGITALomnibus LENS: The alleged failure stemmed from insufficient Decision Visibility across the interconnected systems of kitchen operations, AI dispatch logic, and third-party delivery partners.

CASE 006AUSTRALIAN GOVERNMENT
Read Case Study

Robodebt Scandal — The Catastrophic Cost of Poor Decision Reconstruction

The Australian Government introduced an automated debt recovery system (dubbed Robodebt) that used data matching between Centrelink and the Australian Taxation Office to identify and pursue alleged overpayments.

DIGITALomnibus LENS: The failure was not in a single system, but in the inability to explain and justify the decisions that emerged across interconnected government databases.

CASE 007UNITEDHEALTH
Read Case Study

UnitedHealth AI Claims Denials — The Risks of Black Box Decision Making

UnitedHealth Group, one of the largest health insurers in the United States, reportedly deployed an AI system to review and deny medical claims.

DIGITALomnibus LENS: This case exemplifies the dangers of insufficient Decision Visibility when AI systems influence consequential outcomes across complex data ecosystems.

CASE 008ZILLOW
Read Case Study

Zillow iBuying — When Predictive AI Decisions Lack Visibility

Zillow launched an iBuying programme using sophisticated algorithms to predict home values, purchase properties, and resell them for profit.

DIGITALomnibus LENS: This was a failure of Decision Visibility at scale — the inability to fully understand and explain the logic behind thousands of automated purchasing decisions.

REFERENCEDECISION MAP
View the Map

The Digital Decision Map

When a digital decision is challenged, where do you go next? The complete map of people, data, AI, evidence, regulation and outcomes.

A digital decision is challenged.

Can you explain it? Can you prove it?

Could your organisation explain these decisions
if they happened tomorrow?

DigitalOmnibus reviews one critical business decision and tells you whether it can be explained, evidenced and reconstructed if challenged.

You don't need to become an expert in every regulation. You need confidence that important decisions can still be explained.

Find Out If You Are Exposed

Fixed pricing from €495 · Decision Visibility Report · No ongoing commitment

TECHNICAL VERIFICATION

Want to understand whether your organisation could reconstruct a decision?

Our sister platform NIS2Ireland.com provides a complimentary evidence and visibility assessment.

Learn more at NIS2Ireland.com