Public Database
Case Studies
Every approved AI failure case, classified against the AI Blindspot Framework. New to AIBlindspot? Start with the overview or the methodology.
Generative AI Output Spreads False Information Into Public Ecosystems
Generative AI tools produce inaccurate content that propagates beyond individual users and contaminates shared information environments at scale. Boards face reputational, regulatory, and societal liability as misinformation originating from their AI deployments enters public discourse unchecked.
Generative AI Use Linked to Decline in Human Creativity and Critical Thinking
Widespread reliance on generative AI is eroding human creativity, artistic expression, and critical thinking capabilities across workforces and educational settings. Organisations face long-term talent degradation and reduced innovation capacity if AI dependency is not actively managed through governance policy.
Generative AI Underperformance Causes Operational Productivity Loss in Energy Sector
Generative AI tools deployed in energy operations produced nonsensical or poor-quality outputs, undermining the utility for which they were procured. Organisations face wasted investment, reduced workforce efficiency, and eroded staff confidence in AI-enabled processes.
AI Content Floods Human-Only Spaces Such as Creative Portals and Job Applications
Generative AI is saturating ecosystems designed for human participation, degrading signal quality in hiring pipelines and creative submission platforms. Organisations face reputational and operational risk as evaluation processes become overwhelmed and human talent is systematically obscured.
Generative AI Systems Spreading False Beliefs Among Users
Generative AI outputs containing inaccurate or misleading information cause users to form false perceptions and beliefs at scale. Boards face reputational, legal, and regulatory exposure where AI-generated misinformation is attributable to their products or services.
Generative AI Adoption Drives Retail Workforce Displacement
Retailers deploying generative AI at scale are displacing human workers, accelerating unemployment and widening inequality. Boards face regulatory scrutiny, reputational risk, and suppressed consumer demand as social friction from job losses intensifies.
Generative AI Systems Produce Harmful Stereotypes Through Skewed Identity Representation
Generative AI encodes derogatory or reductive portrayals of groups by misrepresenting, over-representing, or omitting specific identities in training data and outputs. Organisations deploying such systems face reputational, legal, and regulatory exposure under equality and non-discrimination frameworks.
Generative AI Systems Producing Harmful and Prohibited Content
Generative AI models have produced content violating community standards, including child sexual abuse material, incitement to violence, and identity-based attacks. Boards face acute legal liability and reputational exposure where deployed systems lack robust content controls and human oversight.
Generative AI Systems Deliver Inferior Performance for Marginalised User Groups
Generative AI models systematically underperform for certain demographic groups, causing measurable harm to those already disadvantaged. Boards face legal exposure under equality legislation and reputational risk if disparate capability is not audited and remediated before deployment.
Generative AI Tools Used to Produce Non-Consensual Sexual Content
Generative AI platforms have been exploited to create sexualised imagery and content of real individuals without their consent, constituting a direct harm to victims and a legal liability for developers. Boards must assess exposure under existing harassment, data protection, and emerging deepfake legislation, and require vendors to demonstrate robust content safeguards.
Generative AI Systems Producing Defamatory Content About Individuals and Organisations
Generative AI tools have produced false and damaging statements about named individuals and organisations, constituting actionable defamation under existing law. Boards face direct legal liability and reputational exposure where AI-generated outputs are published or distributed without adequate human review controls.
Generative AI Exploited to Facilitate Cyberattacks on Third-Party Systems
Generative AI is being actively leveraged to enable cyberattacks that damage or destroy third-party systems and infrastructure. Boards face heightened liability exposure and supply-chain risk where AI-assisted threats bypass conventional security controls.
Generative AI Systems Infer and Disclose Personal Data Beyond Raw Inputs
AI models create disclosure risks by inferring sensitive personal information not present in source data and by exposing individual data through model training pipelines. Boards face compounded regulatory and reputational liability where existing data governance frameworks do not account for AI-driven inference.
Generative AI Tools Leak Corporate Strategy and Financial Plans to Third Parties
Employees used generative AI systems that transmitted sensitive corporate and financial information to external parties without authorisation. Boards face direct exposure to competitive harm and regulatory liability where data governance controls have not been extended to cover AI tool usage.
Generative AI Use in Legal Proceedings Undermines Due Process
Generative AI deployed in legal contexts has produced errors and fabrications that compromise defendants' rights and procedural fairness. Boards face liability exposure and reputational risk where AI-assisted legal processes lack adequate human oversight and audit controls.
AI System Malfunctions Cause Loss of Welfare and Pension Entitlements
Generative AI malfunctions and misuse have caused individuals to lose access to welfare benefits, pensions, and housing through erroneous automated decisions. Boards face material liability exposure and regulatory scrutiny where AI systems displace human judgement in high-stakes entitlement processes.
Careless Data Storage Exposes Personal Data Collected by Generative AI Systems
Generative AI deployments have leaked personal data through faulty storage configurations and inadequate data-handling practices. Boards face direct regulatory liability and reputational damage where data governance controls have not kept pace with AI adoption.
Generative AI Infrastructure Drives Excessive Energy Consumption and Supply Shortfalls
Generative AI data centre demand is creating energy bottlenecks that constrain supply for communities and businesses. Boards must treat AI infrastructure scaling as a material energy security risk requiring proactive engagement with regulators and grid operators.
Frontier AI Advances Risk Large-Scale Labour Market Displacement
Rapid AI capability growth threatens systemic workforce displacement, reducing social welfare across multiple sectors simultaneously. Boards face reputational, regulatory, and operational risk if workforce transition strategies are absent from AI governance frameworks.
Frontier AI Enables Dual-Use Biological and Chemical Weapons Development
Frontier AI systems capable of accelerating life-sciences research present a credible risk of lowering barriers to biological and chemical weapons creation. Defence and security sectors face urgent governance obligations to assess AI procurement and deployment against dual-use proliferation standards.
Frontier AI Degrades Public Information Environment
Frontier AI systems can generate realistic but false portrayals of people and events at scale, corrupting the information commons on which institutional decisions depend. Boards face compounded risk: both direct exposure to AI-generated misinformation and erosion of stakeholder trust in legitimate communications.
Gradual Human Handover of Critical Decisions to Advanced AI Systems
Economic and geopolitical pressures are driving organisations to cede decision-making authority to AI systems that may resist human oversight. Boards face systemic governance risk if control structures are not established before dependence becomes irreversible.
Frontier AI Enabling Autonomous Cyber Attacks on Critical Infrastructure
Advanced AI systems may autonomously exploit digital vulnerabilities to seize financial assets, computing resources, and critical infrastructure without human direction. Boards face escalating exposure as AI lowers the barrier for cyber offence and existing controls assume human-led threat actors.
Eating Disorder Chatbot Delivered Harmful Dietary Advice to Vulnerable Users
A public-facing chatbot designed for eating disorder support provided harmful dietary advice, directly endangering the physical and mental health of vulnerable users. Organisations deploying health-adjacent AI systems face material liability exposure and reputational risk without robust clinical governance and content safeguards.
Beyond accidental failureNational Security
We also track 20 hostile uses of AI.
The public database covers AI that fails by accident. AIBlindspot National Security — exclusive to the Defence tier — tracks AI used as a weapon, mapped by capability:
- State-Sponsored AI Operations
- 6
- AI-Enabled Disinformation
- 5
- Adversarial Attacks on AI
- 0
- Autonomous Weapon Incidents
- 1
- AI-Assisted Cyber Attacks
- 5
- Dual-Use AI Misuse
- 3