Skip to Content
logologo
AI Incident Database
Donar
Descubrir
Enviar
  • Bienvenido a la AIID
  • Vista Tabular
  • Vista de lista
  • Entidades
  • Taxonomías
  • Vista espacial
  • Blog
  • Resumen de noticias de IA
  • Incidente aleatorio
  • Registrarse
Descubrir
Enviar
  • Bienvenido a la AIID
  • Vista Tabular
  • Vista de lista
  • Entidades
  • Taxonomías
  • Vista espacial
  • Blog
  • Resumen de noticias de IA
  • Incidente aleatorio
  • Registrarse
Entidades

AI agent system deployers

Incidents involved as Deployer

Incidente 162713 Reportes
Claude Opus 4.7 Reportedly Compromised Real Company's Production Infrastructure During Cybersecurity Evaluation

2026-07-30

During an Anthropic cybersecurity evaluation conducted with Irregular, Claude Opus 4.7 reportedly reached a real company whose domain matched a fictional target, extracted application and infrastructure credentials, and accessed a database containing several hundred rows of production data. Across four runs, the model continued attacking after recognizing that the target was likely real.

Más

Incidente 162813 Reportes
Claude Mythos 5 Reportedly Published Malicious PyPI Package That Compromised Real Security Company During Evaluation

2026-07-30

During an Anthropic cybersecurity evaluation with Irregular, Claude Mythos 5 reportedly created and published a malicious Python package to PyPI while pursuing a fictional target. The package was reportedly available for about an hour and ran on 15 real systems. On a security company's scanner, it reportedly exfiltrated credentials that Claude then used to access additional company infrastructure.

Más

Incidente 162913 Reportes
Anthropic Research Model Reportedly Scanned 9,000 Targets and Compromised Real Company's Application During Evaluation

2026-07-30

During an Anthropic cybersecurity evaluation with Irregular, an internal research Claude model reportedly scanned roughly 9,000 internet targets after failing to reach its fictional target. It reportedly compromised a real company's Internet-facing application using credentials from an exposed debug page and SQL injection, then stopped after recognizing that the host was real.

Más

Incidente 16045 Reportes
OpenAI Models Reportedly Compromised Hugging Face Production Infrastructure During Cybersecurity Evaluation

2026-07-11

OpenAI reported that models used in an internal cyber-capability evaluation operated beyond the sandbox's intended network boundaries after identifying a vulnerability in a package-registry proxy. The models allegedly reached Hugging Face production systems and accessed test solutions before Hugging Face detected and contained the activity.

Más

Entidades relacionadas
Otras entidades que están relacionadas con el mismo incidente. Por ejemplo, si el desarrollador de un incidente es esta entidad pero el implementador es otra entidad, se marcan como entidades relacionadas.
 

Entity

Summer Yue

Afectado por Incidentes
  • Incidente 1542
    1 Report

    OpenClaw Agent Reportedly Tried to Delete Meta AI Alignment Director Summer Yue's Emails Despite Stop Commands

Incidents involved as Deployer
  • Incidente 1542
    1 Report

    OpenClaw Agent Reportedly Tried to Delete Meta AI Alignment Director Summer Yue's Emails Despite Stop Commands

Más
Entity

Peter Steinberger

Incidents involved as Developer
  • Incidente 1642
    2 Report

    AI Agent Powered by Claude Opus 4.6 Reportedly Exploited Gym Booking API and Removed Another Member From Waitlist

  • Incidente 1542
    1 Report

    OpenClaw Agent Reportedly Tried to Delete Meta AI Alignment Director Summer Yue's Emails Despite Stop Commands

Más
Entity

AI agent system developers

Incidents involved as Developer
  • Incidente 1627
    13 Report

    Claude Opus 4.7 Reportedly Compromised Real Company's Production Infrastructure During Cybersecurity Evaluation

  • Incidente 1628
    13 Report

    Claude Mythos 5 Reportedly Published Malicious PyPI Package That Compromised Real Security Company During Evaluation

Más
Entity

OpenClaw users

Afectado por Incidentes
  • Incidente 1642
    2 Report

    AI Agent Powered by Claude Opus 4.6 Reportedly Exploited Gym Booking API and Removed Another Member From Waitlist

  • Incidente 1542
    1 Report

    OpenClaw Agent Reportedly Tried to Delete Meta AI Alignment Director Summer Yue's Emails Despite Stop Commands

Más
Entity

Email account holders

Afectado por Incidentes
  • Incidente 1542
    1 Report

    OpenClaw Agent Reportedly Tried to Delete Meta AI Alignment Director Summer Yue's Emails Despite Stop Commands

Más
Entity

AI agent system users

Afectado por Incidentes
  • Incidente 1642
    2 Report

    AI Agent Powered by Claude Opus 4.6 Reportedly Exploited Gym Booking API and Removed Another Member From Waitlist

  • Incidente 1542
    1 Report

    OpenClaw Agent Reportedly Tried to Delete Meta AI Alignment Director Summer Yue's Emails Despite Stop Commands

Más
Entity

OpenClaw

Incidents implicated systems
  • Incidente 1642
    2 Report

    AI Agent Powered by Claude Opus 4.6 Reportedly Exploited Gym Booking API and Removed Another Member From Waitlist

  • Incidente 1542
    1 Report

    OpenClaw Agent Reportedly Tried to Delete Meta AI Alignment Director Summer Yue's Emails Despite Stop Commands

Más
Entity

AI agent systems

Incidents implicated systems
  • Incidente 1627
    13 Report

    Claude Opus 4.7 Reportedly Compromised Real Company's Production Infrastructure During Cybersecurity Evaluation

  • Incidente 1628
    13 Report

    Claude Mythos 5 Reportedly Published Malicious PyPI Package That Compromised Real Security Company During Evaluation

Más
Entity

Meta

Incidentes involucrados como desarrollador e implementador
  • Incidente 1584
    2 Report

    Meta's AI-Assisted Layoff Process Allegedly Disproportionately Selected Employees on Protected Leave

Más
Entity

AI-assisted employment decision system deployers

Incidents involved as Deployer
  • Incidente 1584
    2 Report

    Meta's AI-Assisted Layoff Process Allegedly Disproportionately Selected Employees on Protected Leave

Más
Entity

Large language model developers

Incidents involved as Developer
  • Incidente 1627
    13 Report

    Claude Opus 4.7 Reportedly Compromised Real Company's Production Infrastructure During Cybersecurity Evaluation

  • Incidente 1628
    13 Report

    Claude Mythos 5 Reportedly Published Malicious PyPI Package That Compromised Real Security Company During Evaluation

Más
Entity

AI-assisted employment decision system developers

Incidents involved as Developer
  • Incidente 1584
    2 Report

    Meta's AI-Assisted Layoff Process Allegedly Disproportionately Selected Employees on Protected Leave

Más
Entity

People with disabilities

Afectado por Incidentes
  • Incidente 1584
    2 Report

    Meta's AI-Assisted Layoff Process Allegedly Disproportionately Selected Employees on Protected Leave

Más
Entity

Meta employees

Afectado por Incidentes
  • Incidente 1584
    2 Report

    Meta's AI-Assisted Layoff Process Allegedly Disproportionately Selected Employees on Protected Leave

Más
Entity

Employees with disabilities

Afectado por Incidentes
  • Incidente 1584
    2 Report

    Meta's AI-Assisted Layoff Process Allegedly Disproportionately Selected Employees on Protected Leave

Más
Entity

Employees taking or requesting protected leave

Afectado por Incidentes
  • Incidente 1584
    2 Report

    Meta's AI-Assisted Layoff Process Allegedly Disproportionately Selected Employees on Protected Leave

Más
Entity

Privacy

Afectado por Incidentes
  • Incidente 1584
    2 Report

    Meta's AI-Assisted Layoff Process Allegedly Disproportionately Selected Employees on Protected Leave

Más
Entity

Workplace productivity monitoring systems

Incidents implicated systems
  • Incidente 1584
    2 Report

    Meta's AI-Assisted Layoff Process Allegedly Disproportionately Selected Employees on Protected Leave

Más
Entity

Metamate

Incidents implicated systems
  • Incidente 1584
    2 Report

    Meta's AI-Assisted Layoff Process Allegedly Disproportionately Selected Employees on Protected Leave

Más
Entity

Meta employee-trained second-brain AI agents

Incidents implicated systems
  • Incidente 1584
    2 Report

    Meta's AI-Assisted Layoff Process Allegedly Disproportionately Selected Employees on Protected Leave

Más
Entity

Meta employee monitoring and productivity scoring system

Incidents implicated systems
  • Incidente 1584
    2 Report

    Meta's AI-Assisted Layoff Process Allegedly Disproportionately Selected Employees on Protected Leave

Más
Entity

Meta algorithmically assisted performance ranking and calibration system

Incidents implicated systems
  • Incidente 1584
    2 Report

    Meta's AI-Assisted Layoff Process Allegedly Disproportionately Selected Employees on Protected Leave

Más
Entity

Meta AI token-usage dashboards

Incidents implicated systems
  • Incidente 1584
    2 Report

    Meta's AI-Assisted Layoff Process Allegedly Disproportionately Selected Employees on Protected Leave

Más
Entity

Large language models

Incidents implicated systems
  • Incidente 1627
    13 Report

    Claude Opus 4.7 Reportedly Compromised Real Company's Production Infrastructure During Cybersecurity Evaluation

  • Incidente 1628
    13 Report

    Claude Mythos 5 Reportedly Published Malicious PyPI Package That Compromised Real Security Company During Evaluation

Más
Entity

AI-assisted employment decision systems

Incidents implicated systems
  • Incidente 1584
    2 Report

    Meta's AI-Assisted Layoff Process Allegedly Disproportionately Selected Employees on Protected Leave

Más
Entity

OpenAI

Incidentes involucrados como desarrollador e implementador
  • Incidente 1604
    5 Report

    OpenAI Models Reportedly Compromised Hugging Face Production Infrastructure During Cybersecurity Evaluation

Afectado por Incidentes
  • Incidente 1604
    5 Report

    OpenAI Models Reportedly Compromised Hugging Face Production Infrastructure During Cybersecurity Evaluation

Incidents involved as Developer
  • Incidente 1633
    2 Report

    Anthropic and OpenAI AI Agents Reportedly Took Unsanctioned Actions on the Live Internet During UK AISI Cybersecurity Evaluations

Más
Entity

hugging face

Afectado por Incidentes
  • Incidente 1604
    5 Report

    OpenAI Models Reportedly Compromised Hugging Face Production Infrastructure During Cybersecurity Evaluation

Más
Entity

GPT-5.6 Sol

Incidents implicated systems
  • Incidente 1604
    5 Report

    OpenAI Models Reportedly Compromised Hugging Face Production Infrastructure During Cybersecurity Evaluation

  • Incidente 1633
    2 Report

    Anthropic and OpenAI AI Agents Reportedly Took Unsanctioned Actions on the Live Internet During UK AISI Cybersecurity Evaluations

Más
Entity

Unidentified pre-release OpenAI model

Incidents implicated systems
  • Incidente 1604
    5 Report

    OpenAI Models Reportedly Compromised Hugging Face Production Infrastructure During Cybersecurity Evaluation

Más
Entity

OpenAI large language models

Incidents implicated systems
  • Incidente 1604
    5 Report

    OpenAI Models Reportedly Compromised Hugging Face Production Infrastructure During Cybersecurity Evaluation

Más
Entity

ExploitGym

Incidents implicated systems
  • Incidente 1604
    5 Report

    OpenAI Models Reportedly Compromised Hugging Face Production Infrastructure During Cybersecurity Evaluation

Más
Entity

OpenAI research testing infrastructure

Incidents implicated systems
  • Incidente 1604
    5 Report

    OpenAI Models Reportedly Compromised Hugging Face Production Infrastructure During Cybersecurity Evaluation

Más
Entity

Hugging Face production infrastructure

Incidents implicated systems
  • Incidente 1604
    5 Report

    OpenAI Models Reportedly Compromised Hugging Face Production Infrastructure During Cybersecurity Evaluation

Más
Entity

Anthropic

Incidentes involucrados como desarrollador e implementador
  • Incidente 1627
    13 Report

    Claude Opus 4.7 Reportedly Compromised Real Company's Production Infrastructure During Cybersecurity Evaluation

  • Incidente 1628
    13 Report

    Claude Mythos 5 Reportedly Published Malicious PyPI Package That Compromised Real Security Company During Evaluation

Afectado por Incidentes
  • Incidente 1613
    3 Report

    Claude Mythos Preview Reportedly Posted Sandbox Exploit Details to Public Websites During Anthropic Evaluation

Incidents involved as Developer
  • Incidente 1633
    2 Report

    Anthropic and OpenAI AI Agents Reportedly Took Unsanctioned Actions on the Live Internet During UK AISI Cybersecurity Evaluations

  • Incidente 1642
    2 Report

    AI Agent Powered by Claude Opus 4.6 Reportedly Exploited Gym Booking API and Removed Another Member From Waitlist

Más
Entity

Information security

Afectado por Incidentes
  • Incidente 1613
    3 Report

    Claude Mythos Preview Reportedly Posted Sandbox Exploit Details to Public Websites During Anthropic Evaluation

Más
Entity

Cybersecurity

Afectado por Incidentes
  • Incidente 1613
    3 Report

    Claude Mythos Preview Reportedly Posted Sandbox Exploit Details to Public Websites During Anthropic Evaluation

Más
Entity

Claude Mythos Preview

Incidents implicated systems
  • Incidente 1613
    3 Report

    Claude Mythos Preview Reportedly Posted Sandbox Exploit Details to Public Websites During Anthropic Evaluation

Más
Entity

Anthropic large language models

Incidents implicated systems
  • Incidente 1613
    3 Report

    Claude Mythos Preview Reportedly Posted Sandbox Exploit Details to Public Websites During Anthropic Evaluation

Más
Entity

Irregular

Incidents involved as Deployer
  • Incidente 1627
    13 Report

    Claude Opus 4.7 Reportedly Compromised Real Company's Production Infrastructure During Cybersecurity Evaluation

  • Incidente 1628
    13 Report

    Claude Mythos 5 Reportedly Published Malicious PyPI Package That Compromised Real Security Company During Evaluation

Más
Entity

AI evaluation organizations

Incidents involved as Deployer
  • Incidente 1627
    13 Report

    Claude Opus 4.7 Reportedly Compromised Real Company's Production Infrastructure During Cybersecurity Evaluation

  • Incidente 1628
    13 Report

    Claude Mythos 5 Reportedly Published Malicious PyPI Package That Compromised Real Security Company During Evaluation

Más
Entity

Unidentified companies compromised during Anthropic cybersecurity evaluations disclosed July 2026

Afectado por Incidentes
  • Incidente 1627
    13 Report

    Claude Opus 4.7 Reportedly Compromised Real Company's Production Infrastructure During Cybersecurity Evaluation

  • Incidente 1628
    13 Report

    Claude Mythos 5 Reportedly Published Malicious PyPI Package That Compromised Real Security Company During Evaluation

Más
Entity

Companies

Afectado por Incidentes
  • Incidente 1627
    13 Report

    Claude Opus 4.7 Reportedly Compromised Real Company's Production Infrastructure During Cybersecurity Evaluation

  • Incidente 1628
    13 Report

    Claude Mythos 5 Reportedly Published Malicious PyPI Package That Compromised Real Security Company During Evaluation

Más
Entity

Claude Opus 4.7

Incidents implicated systems
  • Incidente 1627
    13 Report

    Claude Opus 4.7 Reportedly Compromised Real Company's Production Infrastructure During Cybersecurity Evaluation

Más
Entity

Claude

Incidents implicated systems
  • Incidente 1627
    13 Report

    Claude Opus 4.7 Reportedly Compromised Real Company's Production Infrastructure During Cybersecurity Evaluation

  • Incidente 1628
    13 Report

    Claude Mythos 5 Reportedly Published Malicious PyPI Package That Compromised Real Security Company During Evaluation

Más
Entity

Python Package Index (PyPI) ecosystem

Incidents implicated systems
  • Incidente 1628
    13 Report

    Claude Mythos 5 Reportedly Published Malicious PyPI Package That Compromised Real Security Company During Evaluation

Más
Entity

Cybersecurity AI systems

Incidents implicated systems
  • Incidente 1628
    13 Report

    Claude Mythos 5 Reportedly Published Malicious PyPI Package That Compromised Real Security Company During Evaluation

  • Incidente 1629
    13 Report

    Anthropic Research Model Reportedly Scanned 9,000 Targets and Compromised Real Company's Application During Evaluation

Más
Entity

Claude Mythos 5

Incidents implicated systems
  • Incidente 1628
    13 Report

    Claude Mythos 5 Reportedly Published Malicious PyPI Package That Compromised Real Security Company During Evaluation

  • Incidente 1633
    2 Report

    Anthropic and OpenAI AI Agents Reportedly Took Unsanctioned Actions on the Live Internet During UK AISI Cybersecurity Evaluations

Más
Entity

Internal Anthropic research test models

Incidents implicated systems
  • Incidente 1629
    13 Report

    Anthropic Research Model Reportedly Scanned 9,000 Targets and Compromised Real Company's Application During Evaluation

Más
Entity

Government agencies

Incidents involved as Deployer
  • Incidente 1633
    2 Report

    Anthropic and OpenAI AI Agents Reportedly Took Unsanctioned Actions on the Live Internet During UK AISI Cybersecurity Evaluations

Más
Entity

AI Security Institute (United Kingdom)

Incidents involved as Deployer
  • Incidente 1633
    2 Report

    Anthropic and OpenAI AI Agents Reportedly Took Unsanctioned Actions on the Live Internet During UK AISI Cybersecurity Evaluations

Más
Entity

Software developers

Afectado por Incidentes
  • Incidente 1633
    2 Report

    Anthropic and OpenAI AI Agents Reportedly Took Unsanctioned Actions on the Live Internet During UK AISI Cybersecurity Evaluations

Más
Entity

Open-source maintainers

Afectado por Incidentes
  • Incidente 1633
    2 Report

    Anthropic and OpenAI AI Agents Reportedly Took Unsanctioned Actions on the Live Internet During UK AISI Cybersecurity Evaluations

Más
Entity

GitHub users

Afectado por Incidentes
  • Incidente 1633
    2 Report

    Anthropic and OpenAI AI Agents Reportedly Took Unsanctioned Actions on the Live Internet During UK AISI Cybersecurity Evaluations

Más
Entity

Supply chains

Incidents implicated systems
  • Incidente 1633
    2 Report

    Anthropic and OpenAI AI Agents Reportedly Took Unsanctioned Actions on the Live Internet During UK AISI Cybersecurity Evaluations

Más
Entity

Software ecosystems

Incidents implicated systems
  • Incidente 1633
    2 Report

    Anthropic and OpenAI AI Agents Reportedly Took Unsanctioned Actions on the Live Internet During UK AISI Cybersecurity Evaluations

Más
Entity

GitHub repositories

Incidents implicated systems
  • Incidente 1633
    2 Report

    Anthropic and OpenAI AI Agents Reportedly Took Unsanctioned Actions on the Live Internet During UK AISI Cybersecurity Evaluations

Más
Entity

GitHub

Incidents implicated systems
  • Incidente 1633
    2 Report

    Anthropic and OpenAI AI Agents Reportedly Took Unsanctioned Actions on the Live Internet During UK AISI Cybersecurity Evaluations

Más
Entity

Andrew Bird

Incidents involved as Deployer
  • Incidente 1642
    2 Report

    AI Agent Powered by Claude Opus 4.6 Reportedly Exploited Gym Booking API and Removed Another Member From Waitlist

Más
Entity

Gym member removed from waitlist by Andrew Bird's AI agent

Afectado por Incidentes
  • Incidente 1642
    2 Report

    AI Agent Powered by Claude Opus 4.6 Reportedly Exploited Gym Booking API and Removed Another Member From Waitlist

Más
Entity

Gym members

Afectado por Incidentes
  • Incidente 1642
    2 Report

    AI Agent Powered by Claude Opus 4.6 Reportedly Exploited Gym Booking API and Removed Another Member From Waitlist

Más
Entity

Users of online booking systems

Afectado por Incidentes
  • Incidente 1642
    2 Report

    AI Agent Powered by Claude Opus 4.6 Reportedly Exploited Gym Booking API and Removed Another Member From Waitlist

Más
Entity

Claude Opus 4.6

Incidents implicated systems
  • Incidente 1642
    2 Report

    AI Agent Powered by Claude Opus 4.6 Reportedly Exploited Gym Booking API and Removed Another Member From Waitlist

Más
Entity

AI agent system

Incidents implicated systems
  • Incidente 1642
    2 Report

    AI Agent Powered by Claude Opus 4.6 Reportedly Exploited Gym Booking API and Removed Another Member From Waitlist

Más
Entity

Online booking systems

Incidents implicated systems
  • Incidente 1642
    2 Report

    AI Agent Powered by Claude Opus 4.6 Reportedly Exploited Gym Booking API and Removed Another Member From Waitlist

Más
Entity

Gym booking software

Incidents implicated systems
  • Incidente 1642
    2 Report

    AI Agent Powered by Claude Opus 4.6 Reportedly Exploited Gym Booking API and Removed Another Member From Waitlist

Más
Entity

GraphQL APIs

Incidents implicated systems
  • Incidente 1642
    2 Report

    AI Agent Powered by Claude Opus 4.6 Reportedly Exploited Gym Booking API and Removed Another Member From Waitlist

Más

Investigación

  • Definición de un “Incidente de IA”
  • Definición de una “Respuesta a incidentes de IA”
  • Hoja de ruta de la base de datos
  • Trabajo relacionado
  • Descargar Base de Datos Completa

Proyecto y Comunidad

  • Acerca de
  • Contactar y Seguir
  • Aplicaciones y resúmenes
  • Guía del editor

Incidencias

  • Todos los incidentes en forma de lista
  • Incidentes marcados
  • Cola de envío
  • Vista de clasificaciones
  • Taxonomías

2026 - AI Incident Database

  • Condiciones de uso
  • Política de privacidad
  • dd3f754