Skip to Content
logologo
AI Incident Database
Open TwitterOpen RSS FeedOpen FacebookOpen LinkedInOpen GitHub
Open Menu
Découvrir
Envoyer
  • Bienvenue sur AIID
  • Découvrir les incidents
  • Vue spatiale
  • Vue de tableau
  • Vue de liste
  • Entités
  • Taxonomies
  • Soumettre des rapports d'incident
  • Classement des reporters
  • Blog
  • Résumé de l’Actualité sur l’IA
  • Contrôle des risques
  • Incident au hasard
  • S'inscrire
Fermer
Découvrir
Envoyer
  • Bienvenue sur AIID
  • Découvrir les incidents
  • Vue spatiale
  • Vue de tableau
  • Vue de liste
  • Entités
  • Taxonomies
  • Soumettre des rapports d'incident
  • Classement des reporters
  • Blog
  • Résumé de l’Actualité sur l’IA
  • Contrôle des risques
  • Incident au hasard
  • S'inscrire
Fermer

Incident 65: Reinforcement Learning Reward Functions in Video Games

Description: OpenAI published a post about its findings when using Universe, a software for measuring and training AI agents to conduct reinforcement learning experiments, showing that the AI agent did not act in the way intended to complete a videogame.

Outils

Nouveau rapportNouveau rapportNouvelle RéponseNouvelle RéponseDécouvrirDécouvrirVoir l'historiqueVoir l'historique

Entités

Voir toutes les entités
Présumé : Un système d'IA développé et mis en œuvre par OpenAI, a endommagé OpenAI.

Statistiques d'incidents

ID
65
Nombre de rapports
1
Date de l'incident
2016-12-22
Editeurs
Sean McGregor
Applied Taxonomies
CSETv0, CSETv1, GMF, MIT

Classifications de taxonomie CSETv0

Détails de la taxonomie

Problem Nature

Indicates which, if any, of the following types of AI failure describe the incident: "Specification," i.e. the system's behavior did not align with the true intentions of its designer, operator, etc; "Robustness," i.e. the system operated unsafely because of features or changes in its environment, or in the inputs the system received; "Assurance," i.e. the system could not be adequately monitored or controlled during operation.
 

Specification

Physical System

Where relevant, indicates whether the AI system(s) was embedded into or tightly associated with specific types of hardware.
 

Software only

Level of Autonomy

The degree to which the AI system(s) functions independently from human intervention. "High" means there is no human involved in the system action execution; "Medium" means the system generates a decision and a human oversees the resulting action; "low" means the system generates decision-support output and a human makes a decision and executes an action.
 

Unclear/unknown

Nature of End User

"Expert" if users with special training or technical expertise were the ones meant to benefit from the AI system(s)’ operation; "Amateur" if the AI systems were primarily meant to benefit the general public or untrained users.
 

Expert

Public Sector Deployment

"Yes" if the AI system(s) involved in the accident were being used by the public sector or for the administration of public goods (for example, public transportation). "No" if the system(s) were being used in the private sector or for commercial purposes (for example, a ride-sharing company), on the other.
 

No

Data Inputs

A brief description of the data that the AI system(s) used or were trained on.
 

Universe software training

Classifications de taxonomie CSETv1

Détails de la taxonomie

Incident Number

The number of the incident in the AI Incident Database.
 

65

Special Interest Intangible Harm

An assessment of whether a special interest intangible harm occurred. This assessment does not consider the context of the intangible harm, if an AI was involved, or if there is characterizable class or subgroup of harmed entities. It is also not assessing if an intangible harm occurred. It is only asking if a special interest intangible harm occurred.
 

no

Date of Incident Year

The year in which the incident occurred. If there are multiple harms or occurrences of the incident, list the earliest. If a precise date is unavailable, but the available sources provide a basis for estimating the year, estimate. Otherwise, leave blank. Enter in the format of YYYY
 

2016

Date of Incident Month

The month in which the incident occurred. If there are multiple harms or occurrences of the incident, list the earliest. If a precise date is unavailable, but the available sources provide a basis for estimating the month, estimate. Otherwise, leave blank. Enter in the format of MM
 

12

Date of Incident Day

The day on which the incident occurred. If a precise date is unavailable, leave blank. Enter in the format of DD
 

21

Estimated Date

“Yes” if the data was estimated. “No” otherwise.
 

Yes

Classifications de taxonomie MIT

Machine-Classified
Détails de la taxonomie

Risk Subdomain

A further 23 subdomains create an accessible and understandable classification of hazards and harms associated with AI
 

7.1. AI pursuing its own goals in conflict with human goals or values

Risk Domain

The Domain Taxonomy of AI Risks classifies risks into seven AI risk domains: (1) Discrimination & toxicity, (2) Privacy & security, (3) Misinformation, (4) Malicious actors & misuse, (5) Human-computer interaction, (6) Socioeconomic & environmental harms, and (7) AI system safety, failures & limitations.
 
  1. AI system safety, failures, and limitations

Entity

Which, if any, entity is presented as the main cause of the risk
 

AI

Timing

The stage in the AI lifecycle at which the risk is presented as occurring
 

Post-deployment

Intent

Whether the risk is presented as occurring as an expected or unexpected outcome from pursuing a goal
 

Unintentional

Rapports d'incidents

Chronologie du rapport

+1
Fonctions de récompense défectueuses dans la nature
Fonctions de récompense défectueuses dans la nature

Fonctions de récompense défectueuses dans la nature

blog.openai.com

Fonctions de récompense défectueuses dans la nature
blog.openai.com · 2016
Traduit par IA

Chez OpenAI, nous avons récemment commencé à utiliser Universe, notre logiciel de mesure et de formation d'agents IA, pour mener de nouvelles expériences RL. Parfois, ces expériences illustrent certains des problèmes de RL tels qu'ils sont …

Variantes

Une "Variante" est un incident qui partage les mêmes facteurs de causalité, produit des dommages similaires et implique les mêmes systèmes intelligents qu'un incident d'IA connu. Plutôt que d'indexer les variantes comme des incidents entièrement distincts, nous listons les variations d'incidents sous le premier incident similaire soumis à la base de données. Contrairement aux autres types de soumission à la base de données des incidents, les variantes ne sont pas tenues d'avoir des rapports en preuve externes à la base de données des incidents. En savoir plus sur le document de recherche.

Incidents similaires

Par similarité textuelle

Did our AI mess up? Flag the unrelated incidents

Biased Sentiment Analysis

Google's sentiment analysis API is just as biased as humans

Oct 2017 · 7 rapports
Gender Biases in Google Translate

Semantics derived automatically from language corpora contain human-like biases

Apr 2017 · 10 rapports
Tesla Autopilot’s Lane Recognition Allegedly Vulnerable to Adversarial Attacks

AI Incident Database Incidents Converted to Issues

Mar 2019 · 1 rapport
Incident précédentProchain incident

Incidents similaires

Par similarité textuelle

Did our AI mess up? Flag the unrelated incidents

Biased Sentiment Analysis

Google's sentiment analysis API is just as biased as humans

Oct 2017 · 7 rapports
Gender Biases in Google Translate

Semantics derived automatically from language corpora contain human-like biases

Apr 2017 · 10 rapports
Tesla Autopilot’s Lane Recognition Allegedly Vulnerable to Adversarial Attacks

AI Incident Database Incidents Converted to Issues

Mar 2019 · 1 rapport

Recherche

  • Définition d'un « incident d'IA »
  • Définir une « réponse aux incidents d'IA »
  • Feuille de route de la base de données
  • Travaux connexes
  • Télécharger la base de données complète

Projet et communauté

  • À propos de
  • Contacter et suivre
  • Applications et résumés
  • Guide de l'éditeur

Incidents

  • Tous les incidents sous forme de liste
  • Incidents signalés
  • File d'attente de soumission
  • Affichage des classifications
  • Taxonomies

2024 - AI Incident Database

  • Conditions d'utilisation
  • Politique de confidentialité
  • Open twitterOpen githubOpen rssOpen facebookOpen linkedin
  • ecd56df