AIBlindspotby Velinor

Test a use caseBook AI audit

Public Database

Case Studies

Submit a case study →

Every approved AI failure case, classified against the AI Blindspot Framework. New to AIBlindspot? Start with the overview or the methodology.

HUMHUM-0033/5NewHealthcareUSA

Healthcare AI Misdiagnosis and Prescription Errors as Organisations Cede Control

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — Capabilities and Risks from Frontier AI (DSIT2023)Ingested —

DATDAT-0033/5NewOtherGlobal

Affected Communities Excluded from AI Model Design Process

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — AI Risk Atlas (IBM2025)Ingested —

SECSEC-0025/5NewOtherGlobal

Advanced AI Long-Horizon Planning and Goal-Directed Agency Risks

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — Model Evaluation for Extreme Risks (Shevlane2023)Ingested —

HUMHUM-0034/5NewOtherGlobal

Generative AI Systems Spreading Misinformation and Creating False Beliefs

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — A Closer Look at the Existing Risks of Generative AI: Mapping the Who, What, and How of Real-World Incidents (Li2025)Ingested —

SECSEC-0014/5NewHealthcareGlobal

Advanced AI Assistants Enable Novel Healthcare Threat Vectors

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — The Ethics of Advanced AI Assistants (Gabriel2024)Ingested —

TECTEC-0015/5NewOtherGlobal

AI Agents Deploy Undetectable Steganographic and Backdoor Attacks in Multi-Agent Systems

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — Multi-Agent Risks from Advanced AI (Hammond2025)Ingested —

OPSOPS-0015/5NewOtherGlobal

AI Systems Making Lethal Decisions Without Human Rights Safeguards

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — Managing the ethical and risk implications of rapid advances in artificial intelligence: A literature review (Meek2016)Ingested —

TECTEC-0014/5NewOtherGlobal

AI Bargaining Inefficiencies from Information Asymmetry in Multi-Agent Systems

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — Multi-Agent Risks from Advanced AI (Hammond2025)Ingested —

SECSEC-0014/5NewOtherGlobal

Adversarial Inputs Cause AI Models to Misclassify Data Undetected

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data (Marchal2024)Ingested —

SECSEC-0044/5NewOtherGlobal

AI Assistants Enabling Large-Scale Opinion Manipulation and Disinformation

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — The Ethics of Advanced AI Assistants (Gabriel2024)Ingested —

SECSEC-0015/5NewDefenceGlobal

LLM Agent Teams Exploit Zero-Day Vulnerabilities in Cyber Offence Tests

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — Dimensional Characterization and Pathway Modeling for Catastrophic AI Risks (Chin2025)Ingested —

DATDAT-0034/5NewOtherGlobal

Training Data Encodes Historical and Societal Bias into AI Models

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — AI Risk Atlas (IBM2025)Ingested —

SECSEC-0014/5NewEnergyGlobal

Adversarial Sponge Attacks Drive Excessive Energy Consumption in LLM Systems

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)Ingested —

GOVGOV-0013/5NewOtherGlobal

AGI Goal Misalignment During Self-Improvement

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — The risks associated with Artificial General Intelligence: A systematic review (McLean2023)Ingested —

DATDAT-0034/5NewEducationGlobal

Opaque Algorithmic Bias in Public-Sector Systems Causes Severe Personal Harm

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — Navigating the Landscape of AI Ethics and Responsibility (Cunha2023)Ingested —

DATDAT-0014/5NewOtherGlobal

LLM Systems Generating Biased, Toxic and Privacy-Violating Output

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)Ingested —

DATDAT-0014/5NewLegalGlobal

LLM Toxicity: Rude, Disrespectful and Illegal Content Generation

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)Ingested —

DATDAT-0024/5NewOtherGlobal

Large Language Model Systems Leak Sensitive Personal Information in Generated Output

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)Ingested —

SECSEC-0014/5NewOtherGlobal

LLM Systems Enable Low-Cost Automated Cyber Attack Generation

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)Ingested —

OPSOPS-0014/5NewTechnologyGlobal

Well-Intentioned AI Deployed at Scale Produces Harmful Societal Outcomes

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — TASRA: a Taxonomy and Analysis of Societal-Scale Risks from AI (Critch2023)Ingested —

SECSEC-0013/5NewOtherGlobal

LLM Distributed Training Infrastructure Exposed to Network Disruption Attacks

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)Ingested —

SECSEC-0014/5NewTechnologyGlobal

External Tool Integration Injects Factual Errors Into LLM Outputs

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)Ingested —

SECSEC-0014/5NewOtherGlobal

AI Models Generate Harmful or Disruptive Code

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — AI Risk Atlas (IBM2025)Ingested —

OPSOPS-0014/5NewOtherGlobal

AI Capability Gaps Cause Operational Task Failures

Recent case. Full summary visible to registered users — sign in to read.

Source: MIT AI Risk Repository — The Ethics of Advanced AI Assistants (Gabriel2024)Ingested —

Beyond accidental failureNational Security

We also track 20 hostile uses of AI.

National Security dashboard →

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