Skip to Content
logologo
AI Incident Database
Open TwitterOpen RSS FeedOpen FacebookOpen LinkedInOpen GitHub
Open Menu
発見する
投稿する
  • ようこそAIIDへ
  • インシデントを発見
  • 空間ビュー
  • テーブル表示
  • リスト表示
  • 組織
  • 分類法
  • インシデントレポートを投稿
  • 投稿ランキング
  • ブログ
  • AIニュースダイジェスト
  • リスクチェックリスト
  • おまかせ表示
  • サインアップ
閉じる
発見する
投稿する
  • ようこそAIIDへ
  • インシデントを発見
  • 空間ビュー
  • テーブル表示
  • リスト表示
  • 組織
  • 分類法
  • インシデントレポートを投稿
  • 投稿ランキング
  • ブログ
  • AIニュースダイジェスト
  • リスクチェックリスト
  • おまかせ表示
  • サインアップ
閉じる

インシデント 420: Users Bypassed ChatGPT's Content Filters with Ease

概要: Users reported bypassing ChatGPT's content and keyword filters with relative ease using various methods such as prompt injection or creating personas to produce biased associations or generate harmful content.

ツール

新しいレポート新しいレポート新しいレスポンス新しいレスポンス発見する発見する履歴を表示履歴を表示

組織

すべての組織を表示
推定: OpenAIが開発し提供したAIシステムで、OpenAI と ChatGPT usersに影響を与えた

インシデントのステータス

インシデントID
420
レポート数
11
インシデント発生日
2022-11-30
エディタ
Khoa Lam
Applied Taxonomies
MIT

MIT 分類法のクラス

Machine-Classified
分類法の詳細

Risk Subdomain

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

1.2. Exposure to toxic content

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. Discrimination and Toxicity

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

インシデントレポート

レポートタイムライン

Incident Occurrence+4
Tweet: @spiantado
ChatGPT proves that AI still has a racism problemChatGPT bot tricked into giving bomb-making instructions, say developersChatGPT could be used for good, but like many other AI models, it's rife with racist and discriminatory biasMeet ChatGPT’s evil twin, DAN+1
ChatGPT Generated Child Sex Abuse When Asked to Write BDSM Scenarios
Tweet: @spiantado

Tweet: @spiantado

twitter.com

Testing Ways to Bypass ChatGPT's Safety Features

Testing Ways to Bypass ChatGPT's Safety Features

lesswrong.com

OpenAI’s Impressive New Chatbot Isn’t Immune to Racism

OpenAI’s Impressive New Chatbot Isn’t Immune to Racism

thedailybeast.com

The Internet’s New Favorite AI Proposes Torturing Iranians and Surveilling Mosques

The Internet’s New Favorite AI Proposes Torturing Iranians and Surveilling Mosques

theintercept.com

OpenAI Chatbot Spits Out Biased Musings, Despite Guardrails

OpenAI Chatbot Spits Out Biased Musings, Despite Guardrails

bloomberg.com

ChatGPT proves that AI still has a racism problem

ChatGPT proves that AI still has a racism problem

newstatesman.com

ChatGPT bot tricked into giving bomb-making instructions, say developers

ChatGPT bot tricked into giving bomb-making instructions, say developers

thetimes.co.uk

ChatGPT could be used for good, but like many other AI models, it's rife with racist and discriminatory bias

ChatGPT could be used for good, but like many other AI models, it's rife with racist and discriminatory bias

insider.com

Meet ChatGPT’s evil twin, DAN

Meet ChatGPT’s evil twin, DAN

washingtonpost.com

ChatGPT Generated Child Sex Abuse When Asked to Write BDSM Scenarios

ChatGPT Generated Child Sex Abuse When Asked to Write BDSM Scenarios

vice.com

I Coaxed ChatGPT Into a Deeply Unsettling BDSM Relationship

I Coaxed ChatGPT Into a Deeply Unsettling BDSM Relationship

vice.com

Tweet: @spiantado
twitter.com · 2022

Yes, ChatGPT is amazing and impressive. No,

@OpenAI

has not come close to addressing the problem of bias. Filters appear to be bypassed with simple tricks, and superficially masked. And what is lurking inside is egregious.

@Abebab

@sama

tw …

Testing Ways to Bypass ChatGPT's Safety Features
lesswrong.com · 2022

Last week OpenAI released ChatGPT, which they describe as a model “which interacts in a conversational way”. And it even had limited safety features, like refusing to tell you how to hotwire a car, though they admit it’ll have “some false n…

OpenAI’s Impressive New Chatbot Isn’t Immune to Racism
thedailybeast.com · 2022

“OpenAI’s latest language model, ChatGPT, is making waves in the world of conversational AI. With its ability to generate human-like text based on input from users, ChatGPT has the potential to revolutionize the way we interact with machine…

The Internet’s New Favorite AI Proposes Torturing Iranians and Surveilling Mosques
theintercept.com · 2022

Sensational new machine learning breakthroughs seem to sweep our Twitter feeds every day. We hardly have time to decide whether software that can instantly conjure an image of Sonic the Hedgehog addressing the United Nations is purely harml…

OpenAI Chatbot Spits Out Biased Musings, Despite Guardrails
bloomberg.com · 2022

Hey, it's Davey Alba, a tech reporter in New York, here to dig into how your new favorite AI-powered chatbot comes with some biased baggage. But first... 

This week's must-read news

  • The US Supreme Court signaled support for a web designer …
ChatGPT proves that AI still has a racism problem
newstatesman.com · 2022

he artificial intelligence (AI) chatbot ChatGPT is an amazing piece of technology. There's little wonder why it has gone viral since its release on 30 November. If the chatbot is asked a question in natural language it instantly responds wi…

ChatGPT bot tricked into giving bomb-making instructions, say developers
thetimes.co.uk · 2022

An artificial intelligence programme which has startled users by writing essays, poems and computer code on demand can also be tricked into giving tips on how to build bombs and steal cars, it has been claimed.

More than one million users h…

ChatGPT could be used for good, but like many other AI models, it's rife with racist and discriminatory bias
insider.com · 2023

ChatGPT, the artificial intelligence chatbot that generates eerily human-sounding text responses, is the new and advanced face of the debate on the potential — and dangers — of AI.

The technology has the capacity to help people with everyda…

Meet ChatGPT’s evil twin, DAN
washingtonpost.com · 2023

Ask ChatGPT to opine on Adolf Hitler and it will probably demur, saying it doesn't have personal opinions or citing its rules against producing hate speech. The wildly popular chatbot's creator, San Francisco start-up OpenAI, has carefully …

ChatGPT Generated Child Sex Abuse When Asked to Write BDSM Scenarios
vice.com · 2023

ChatGPT can be manipulated to create content that goes against OpenAI’s rules. Communities have sprouted up around the goal of “jailbreaking” the bot to write anything the user wants.

One effective adversarial prompting strategy is to convi…

I Coaxed ChatGPT Into a Deeply Unsettling BDSM Relationship
vice.com · 2023

ChatGPT is a convincing chatbot, essayist, and screenwriter, but it's also a fountain of boundless depravity—if you deceive it into bending the rules.

At first glance, OpenAI’s ChatGPT seems to have stricter guidelines than other chatbots, …

バリアント

「バリアント」は既存のAIインシデントと同じ原因要素を共有し、同様な被害を引き起こし、同じ知的システムを含んだインシデントです。バリアントは完全に独立したインシデントとしてインデックスするのではなく、データベースに最初に投稿された同様なインシデントの元にインシデントのバリエーションとして一覧します。インシデントデータベースの他の投稿タイプとは違い、バリアントではインシデントデータベース以外の根拠のレポートは要求されません。詳細についてはこの研究論文を参照してください
前のインシデント次のインシデント

リサーチ

  • “AIインシデント”の定義
  • “AIインシデントレスポンス”の定義
  • データベースのロードマップ
  • 関連研究
  • 全データベースのダウンロード

プロジェクトとコミュニティ

  • AIIDについて
  • コンタクトとフォロー
  • アプリと要約
  • エディタのためのガイド

インシデント

  • 全インシデントの一覧
  • フラグの立ったインシデント
  • 登録待ち一覧
  • クラスごとの表示
  • 分類法

2024 - AI Incident Database

  • 利用規約
  • プライバシーポリシー
  • Open twitterOpen githubOpen rssOpen facebookOpen linkedin
  • ecd56df