Risques et atténuation

{
  "id": "risks_mitigation",
  "name": [
    {
      "locale": "en",
      "value": "Risks & Mitigation"
    },
    {
      "locale": "fr",
      "value": "Risques et atténuation"
    }
  ],
  "description": [
    {
      "locale": "en",
      "value": "The risks this AI system could create and how they’re being addressed."
    },
    {
      "locale": "fr",
      "value": "Les risques que ce système d’IA pourrait créer et la manière dont ils sont traités."
    }
  ],
  "prompt": [
    {
      "locale": "en",
      "value": "What are the risks and how are they mitigated?"
    },
    {
      "locale": "fr",
      "value": "Quels sont les risques et comment sont-ils atténués ?"
    }
  ],
  "authoring_guidance": [],
  "examples": [],
  "sources": [
    {
      "type": "research_paper",
      "title": "A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harms",
      "url": "https://arxiv.org/abs/2407.01294",
      "citation": "Abercrombie et al. (2024). arXiv:2407.01294. CC BY-SA 4.0."
    }
  ],
  "required": false,
  "order": 5,
  "datachain_type": "ai",
  "shape": "octagon",
  "element_variables": [
    {
      "id": "risk",
      "label": [
        {
          "locale": "en",
          "value": "Description of the risk"
        },
        {
          "locale": "fr",
          "value": "Description du risque"
        }
      ],
      "required": true,
      "kind": "localized_text"
    },
    {
      "id": "mitigation",
      "label": [
        {
          "locale": "en",
          "value": "Description of the mitigation"
        },
        {
          "locale": "fr",
          "value": "Description de l'atténuation"
        }
      ],
      "required": false,
      "kind": "localized_text"
    }
  ]
}

[
  {
    "id": "autonomy_loss",
    "category_id": "risks_mitigation",
    "title": [
      {
        "locale": "en",
        "value": "Loss of autonomy"
      },
      {
        "locale": "fr",
        "value": "Perte d'autonomie"
      }
    ],
    "description": [
      {
        "locale": "en",
        "value": "The risk that the AI system limits a person's ability to make their own decisions, control their identity, or choose freely. This includes risks of restricted access to alternatives, deceptive design that nudges decisions toward a particular outcome, and lack of meaningful consent. Mitigations may include opt-out mechanisms, transparent recommendations, defaults that protect agency, and human review of consequential decisions."
      },
      {
        "locale": "fr",
        "value": "Le risque que le système d'IA limite la capacité d'une personne à prendre ses propres décisions, à contrôler son identité ou à choisir librement. Cela inclut les risques d'accès restreint à des alternatives, de conception trompeuse qui oriente les décisions vers un résultat particulier et d'absence de consentement significatif. Les mesures d'atténuation peuvent inclure des mécanismes de retrait, des recommandations transparentes, des paramètres par défaut qui protègent l'agentivité et un examen humain des décisions importantes."
      }
    ],
    "authoring_guidance": [],
    "examples": [],
    "sources": [],
    "symbol_id": "risks_autonomy",
    "variables": [
      {
        "id": "risk",
        "label": [
          {
            "locale": "en",
            "value": "Description of the risk"
          },
          {
            "locale": "fr",
            "value": "Description du risque"
          }
        ],
        "required": true,
        "kind": "localized_text"
      },
      {
        "id": "mitigation",
        "label": [
          {
            "locale": "en",
            "value": "Description of the mitigation"
          },
          {
            "locale": "fr",
            "value": "Description de l'atténuation"
          }
        ],
        "required": false,
        "kind": "localized_text"
      }
    ],
    "shape": "octagon",
    "icon_variants": [
      "default",
      "dark"
    ]
  },
  {
    "id": "civil_liberties_harm",
    "category_id": "risks_mitigation",
    "title": [
      {
        "locale": "en",
        "value": "Civil liberties harm"
      },
      {
        "locale": "fr",
        "value": "Atteinte aux libertés civiles"
      }
    ],
    "description": [
      {
        "locale": "en",
        "value": "The risk that the AI system compromises fundamental human rights and civil liberties — speech, assembly, movement, due process, or freedom from arbitrary surveillance. This includes risks from mass facial recognition, predictive policing, censorship of lawful expression, and chilling effects on protest. Mitigations may include strict purpose limits, judicial or independent oversight, narrow data retention, and human-rights impact assessments before deployment."
      },
      {
        "locale": "fr",
        "value": "Le risque que le système d'IA compromette les droits humains fondamentaux et les libertés civiles — expression, réunion, circulation, procédure équitable ou protection contre la surveillance arbitraire. Cela inclut les risques liés à la reconnaissance faciale de masse, à la police prédictive, à la censure de l'expression légale et aux effets dissuasifs sur la contestation. Les mesures d'atténuation peuvent inclure des limites strictes de finalité, une supervision judiciaire ou indépendante, une rétention de données restreinte et des évaluations d'impact sur les droits humains avant le déploiement."
      }
    ],
    "authoring_guidance": [],
    "examples": [],
    "sources": [],
    "symbol_id": "risks_civil-liberties",
    "variables": [
      {
        "id": "risk",
        "label": [
          {
            "locale": "en",
            "value": "Description of the risk"
          },
          {
            "locale": "fr",
            "value": "Description du risque"
          }
        ],
        "required": true,
        "kind": "localized_text"
      },
      {
        "id": "mitigation",
        "label": [
          {
            "locale": "en",
            "value": "Description of the mitigation"
          },
          {
            "locale": "fr",
            "value": "Description de l'atténuation"
          }
        ],
        "required": false,
        "kind": "localized_text"
      }
    ],
    "shape": "octagon",
    "icon_variants": [
      "default",
      "dark"
    ]
  },
  {
    "id": "environmental_harm",
    "category_id": "risks_mitigation",
    "title": [
      {
        "locale": "en",
        "value": "Environmental harm"
      },
      {
        "locale": "fr",
        "value": "Préjudice environnemental"
      }
    ],
    "description": [
      {
        "locale": "en",
        "value": "The risk that the AI system causes environmental degradation through energy consumption, water use, e-waste, carbon emissions, or resource extraction for hardware. This includes risks from training-time compute, always-on inference, cooling demands of data centers, and rapid hardware churn. Mitigations may include energy and water efficiency targets, transparent reporting of compute footprint, hardware-lifecycle planning, and renewable-energy procurement for data-centre operations."
      },
      {
        "locale": "fr",
        "value": "Le risque que le système d'IA cause une dégradation environnementale par la consommation d'énergie, l'usage de l'eau, les déchets électroniques, les émissions de carbone ou l'extraction de ressources pour le matériel. Cela inclut les risques liés au calcul d'entraînement, à l'inférence permanente, aux besoins de refroidissement des centres de données et au renouvellement rapide du matériel. Les mesures d'atténuation peuvent inclure des objectifs d'efficacité énergétique et hydrique, un reporting transparent de l'empreinte de calcul, une planification du cycle de vie du matériel et l'approvisionnement en énergies renouvelables pour les centres de données."
      }
    ],
    "authoring_guidance": [],
    "examples": [],
    "sources": [],
    "symbol_id": "risks_environmental",
    "variables": [
      {
        "id": "risk",
        "label": [
          {
            "locale": "en",
            "value": "Description of the risk"
          },
          {
            "locale": "fr",
            "value": "Description du risque"
          }
        ],
        "required": true,
        "kind": "localized_text"
      },
      {
        "id": "mitigation",
        "label": [
          {
            "locale": "en",
            "value": "Description of the mitigation"
          },
          {
            "locale": "fr",
            "value": "Description de l'atténuation"
          }
        ],
        "required": false,
        "kind": "localized_text"
      }
    ],
    "shape": "octagon",
    "icon_variants": [
      "default",
      "dark"
    ]
  },
  {
    "id": "financial_harm",
    "category_id": "risks_mitigation",
    "title": [
      {
        "locale": "en",
        "value": "Financial & business harm"
      },
      {
        "locale": "fr",
        "value": "Préjudice financier et commercial"
      }
    ],
    "description": [
      {
        "locale": "en",
        "value": "The risk that the AI system causes financial losses to individuals or organizational damage through erroneous pricing, denied services, fraud, or market manipulation. This includes risks from automated credit decisions, dynamic-pricing inequities, AI-enabled scams, and supply-chain disruption from misuse. Mitigations may include affordability guards, fairness audits of pricing models, fraud-detection layers, and clear redress procedures."
      },
      {
        "locale": "fr",
        "value": "Le risque que le système d'IA cause des pertes financières à des personnes ou des dommages organisationnels par une tarification erronée, des refus de services, des fraudes ou une manipulation du marché. Cela inclut les risques liés aux décisions de crédit automatisées, aux inégalités de tarification dynamique, aux escroqueries activées par l'IA et aux perturbations de la chaîne d'approvisionnement résultant d'un usage abusif. Les mesures d'atténuation peuvent inclure des garde-fous d'accessibilité financière, des audits d'équité des modèles de tarification, des couches de détection de la fraude et des procédures claires de recours."
      }
    ],
    "authoring_guidance": [],
    "examples": [],
    "sources": [],
    "symbol_id": "risks_financial",
    "variables": [
      {
        "id": "risk",
        "label": [
          {
            "locale": "en",
            "value": "Description of the risk"
          },
          {
            "locale": "fr",
            "value": "Description du risque"
          }
        ],
        "required": true,
        "kind": "localized_text"
      },
      {
        "id": "mitigation",
        "label": [
          {
            "locale": "en",
            "value": "Description of the mitigation"
          },
          {
            "locale": "fr",
            "value": "Description de l'atténuation"
          }
        ],
        "required": false,
        "kind": "localized_text"
      }
    ],
    "shape": "octagon",
    "icon_variants": [
      "default",
      "dark"
    ]
  },
  {
    "id": "physical_harm",
    "category_id": "risks_mitigation",
    "title": [
      {
        "locale": "en",
        "value": "Physical harm"
      },
      {
        "locale": "fr",
        "value": "Préjudice physique"
      }
    ],
    "description": [
      {
        "locale": "en",
        "value": "The risk that the AI system's outputs lead to physical injury to individuals or damage to property. This includes risks from autonomous vehicles, robotic actuation, faulty navigation guidance, or critical-infrastructure errors that put bodies or property in harm's way. Mitigations may include rigorous safety testing, fail-safe defaults, geofencing of hazardous behavior, and human oversight of high-stakes physical actions."
      },
      {
        "locale": "fr",
        "value": "Le risque que les résultats du système d'IA entraînent des blessures physiques pour des personnes ou des dommages matériels. Cela inclut les risques liés aux véhicules autonomes, à l'actionnement robotique, à des indications de navigation erronées ou à des erreurs d'infrastructures critiques mettant en danger des corps ou des biens. Les mesures d'atténuation peuvent inclure des tests de sécurité rigoureux, des paramètres par défaut à sécurité intégrée, le géorepérage des comportements dangereux et la supervision humaine des actions physiques à enjeux élevés."
      }
    ],
    "authoring_guidance": [],
    "examples": [],
    "sources": [],
    "symbol_id": "risks_physical",
    "variables": [
      {
        "id": "risk",
        "label": [
          {
            "locale": "en",
            "value": "Description of the risk"
          },
          {
            "locale": "fr",
            "value": "Description du risque"
          }
        ],
        "required": true,
        "kind": "localized_text"
      },
      {
        "id": "mitigation",
        "label": [
          {
            "locale": "en",
            "value": "Description of the mitigation"
          },
          {
            "locale": "fr",
            "value": "Description de l'atténuation"
          }
        ],
        "required": false,
        "kind": "localized_text"
      }
    ],
    "shape": "octagon",
    "icon_variants": [
      "default",
      "dark"
    ]
  },
  {
    "id": "political_economic_harm",
    "category_id": "risks_mitigation",
    "title": [
      {
        "locale": "en",
        "value": "Political & economic harm"
      },
      {
        "locale": "fr",
        "value": "Préjudice politique et économique"
      }
    ],
    "description": [
      {
        "locale": "en",
        "value": "The risk that the AI system manipulates political discourse, interferes with elections, concentrates market power, or damages public institutions. This includes risks of synthetic media targeting voters, opaque ad-targeting, competitive distortion from monopoly access to data and compute, and erosion of institutional trust. Mitigations may include disclosure of synthetic content, antitrust attention to AI markets, audit access for regulators, and political-ad transparency requirements."
      },
      {
        "locale": "fr",
        "value": "Le risque que le système d'IA manipule le discours politique, interfère avec les élections, concentre le pouvoir de marché ou porte atteinte aux institutions publiques. Cela inclut les risques liés aux médias synthétiques ciblant les électeurs, au ciblage publicitaire opaque, à la distorsion concurrentielle due à l'accès monopolistique aux données et au calcul, et à l'érosion de la confiance institutionnelle. Les mesures d'atténuation peuvent inclure la divulgation des contenus synthétiques, l'attention antitrust portée aux marchés de l'IA, l'accès d'audit pour les régulateurs et des obligations de transparence pour les publicités politiques."
      }
    ],
    "authoring_guidance": [],
    "examples": [],
    "sources": [],
    "symbol_id": "risks_political-economic",
    "variables": [
      {
        "id": "risk",
        "label": [
          {
            "locale": "en",
            "value": "Description of the risk"
          },
          {
            "locale": "fr",
            "value": "Description du risque"
          }
        ],
        "required": true,
        "kind": "localized_text"
      },
      {
        "id": "mitigation",
        "label": [
          {
            "locale": "en",
            "value": "Description of the mitigation"
          },
          {
            "locale": "fr",
            "value": "Description de l'atténuation"
          }
        ],
        "required": false,
        "kind": "localized_text"
      }
    ],
    "shape": "octagon",
    "icon_variants": [
      "default",
      "dark"
    ]
  },
  {
    "id": "psychological_harm",
    "category_id": "risks_mitigation",
    "title": [
      {
        "locale": "en",
        "value": "Psychological harm"
      },
      {
        "locale": "fr",
        "value": "Préjudice psychologique"
      }
    ],
    "description": [
      {
        "locale": "en",
        "value": "The risk that the AI system causes emotional or mental-health impairment, directly or indirectly. This includes risks from addictive feedback loops, distress caused by surveillance, harassment via generated content, and anxiety from automated denials. Mitigations may include content warnings, well-being safeguards, age-appropriate design, accessible support channels, and rate limits on engagement-maximizing behavior."
      },
      {
        "locale": "fr",
        "value": "Le risque que le système d'IA cause une atteinte émotionnelle ou à la santé mentale, directement ou indirectement. Cela inclut les risques liés à des boucles de rétroaction addictives, à la détresse causée par la surveillance, au harcèlement via du contenu généré et à l'anxiété liée aux refus automatisés. Les mesures d'atténuation peuvent inclure des avertissements de contenu, des garde-fous pour le bien-être, une conception adaptée à l'âge, des canaux d'assistance accessibles et des limites sur les comportements qui maximisent l'engagement."
      }
    ],
    "authoring_guidance": [],
    "examples": [],
    "sources": [],
    "symbol_id": "risks_psychological",
    "variables": [
      {
        "id": "risk",
        "label": [
          {
            "locale": "en",
            "value": "Description of the risk"
          },
          {
            "locale": "fr",
            "value": "Description du risque"
          }
        ],
        "required": true,
        "kind": "localized_text"
      },
      {
        "id": "mitigation",
        "label": [
          {
            "locale": "en",
            "value": "Description of the mitigation"
          },
          {
            "locale": "fr",
            "value": "Description de l'atténuation"
          }
        ],
        "required": false,
        "kind": "localized_text"
      }
    ],
    "shape": "octagon",
    "icon_variants": [
      "default",
      "dark"
    ]
  },
  {
    "id": "reputational_harm",
    "category_id": "risks_mitigation",
    "title": [
      {
        "locale": "en",
        "value": "Reputational harm"
      },
      {
        "locale": "fr",
        "value": "Atteinte à la réputation"
      }
    ],
    "description": [
      {
        "locale": "en",
        "value": "The risk that the AI system damages the reputation of individuals, groups, or organizations through misidentification, false categorization, or stigmatizing labels. This includes risks of inaccurate face recognition, defamatory generated content, and unfair public scoring that follow people across contexts. Mitigations may include accuracy thresholds before public-facing labels are applied, human review of high-impact identifications, takedown procedures, and clear rights to correction."
      },
      {
        "locale": "fr",
        "value": "Le risque que le système d'IA porte atteinte à la réputation de personnes, de groupes ou d'organisations par une identification erronée, une catégorisation fausse ou des étiquettes stigmatisantes. Cela inclut les risques de reconnaissance faciale inexacte, de contenu diffamatoire généré et de notation publique injuste qui suit les personnes dans différents contextes. Les mesures d'atténuation peuvent inclure des seuils de précision avant l'application d'étiquettes publiques, un examen humain des identifications à fort impact, des procédures de retrait et des droits clairs de rectification."
      }
    ],
    "authoring_guidance": [],
    "examples": [],
    "sources": [],
    "symbol_id": "risks_reputational",
    "variables": [
      {
        "id": "risk",
        "label": [
          {
            "locale": "en",
            "value": "Description of the risk"
          },
          {
            "locale": "fr",
            "value": "Description du risque"
          }
        ],
        "required": true,
        "kind": "localized_text"
      },
      {
        "id": "mitigation",
        "label": [
          {
            "locale": "en",
            "value": "Description of the mitigation"
          },
          {
            "locale": "fr",
            "value": "Description de l'atténuation"
          }
        ],
        "required": false,
        "kind": "localized_text"
      }
    ],
    "shape": "octagon",
    "icon_variants": [
      "default",
      "dark"
    ]
  },
  {
    "id": "societal_cultural_harm",
    "category_id": "risks_mitigation",
    "title": [
      {
        "locale": "en",
        "value": "Societal & cultural harm"
      },
      {
        "locale": "fr",
        "value": "Préjudice sociétal et culturel"
      }
    ],
    "description": [
      {
        "locale": "en",
        "value": "The risk that the AI system harms communities or culture through erosion of trust, loss of linguistic and cultural diversity, or unhealthy dependency on opaque systems. This includes risks of misinformation at scale, homogenization of cultural expression, replacement of local knowledge, and degradation of public-information ecosystems. Mitigations may include multilingual support, partnerships with affected communities, provenance signals on generated content, and investment in local-language and local-context models."
      },
      {
        "locale": "fr",
        "value": "Le risque que le système d'IA porte atteinte aux communautés ou à la culture par l'érosion de la confiance, la perte de diversité linguistique et culturelle ou une dépendance malsaine à des systèmes opaques. Cela inclut les risques de désinformation à grande échelle, d'homogénéisation de l'expression culturelle, de substitution des savoirs locaux et de dégradation des écosystèmes d'information publique. Les mesures d'atténuation peuvent inclure un support multilingue, des partenariats avec les communautés concernées, des signaux de provenance sur les contenus générés et des investissements dans des modèles spécifiques aux langues et aux contextes locaux."
      }
    ],
    "authoring_guidance": [],
    "examples": [],
    "sources": [],
    "symbol_id": "risks_societal-cultural",
    "variables": [
      {
        "id": "risk",
        "label": [
          {
            "locale": "en",
            "value": "Description of the risk"
          },
          {
            "locale": "fr",
            "value": "Description du risque"
          }
        ],
        "required": true,
        "kind": "localized_text"
      },
      {
        "id": "mitigation",
        "label": [
          {
            "locale": "en",
            "value": "Description of the mitigation"
          },
          {
            "locale": "fr",
            "value": "Description de l'atténuation"
          }
        ],
        "required": false,
        "kind": "localized_text"
      }
    ],
    "shape": "octagon",
    "icon_variants": [
      "default",
      "dark"
    ]
  }
]