概要: A network of 171 bot accounts on X are alleged to have used ChatGPT to generate political content supporting Ghana’s New Patriotic Party (NPP) and its presidential candidate, Mahamudu Bawumia, ahead of the December 2024 election. The AI-generated posts reportedly praised Bawumia while spreading disinformation targeting the opposition candidate, John Mahama, of the National Democratic Congress (NDC).
Editor Notes: Reconstructing the timeline of events: (1) February 2024: The network of 171 bot accounts is reported to have become active by regularly generating and disseminating political content favoring Ghana’s New Patriotic Party (NPP) and its presidential candidate, Mahamudu Bawumia. The posts reportedly used hashtags like #Bawumia2024 and #ItIsPossible to amplify pro-NPP messaging while disparaging opposition candidate John Mahama with hashtags like #Mahamaisaliar and #DrunkmaniMahama. (2) October 2024: OpenAI reported disrupting over 20 political influence networks worldwide that used ChatGPT to generate political content, though this network targeting Ghana’s election is reportedly not specifically mentioned in its statement. (3) November 12, 2024: NewsGuard releases research findings identifying the 171 bot accounts as likely using ChatGPT to generate political posts. According to NewsGuard, the bots were found to operate with regimented patterns, posting 10 or more times per day during Ghanaian business hours. (4) December 2024: As the presidential election approached on December 7, the bot accounts continued to be active. NewsGuard shared its findings with X and OpenAI, but only four of the 171 accounts were reported to have faced action, with two suspended and two restricted. (8) The Wikipedia article on this election is located at https://en.wikipedia.org/wiki/2024_Ghanaian_general_election.
Alleged: OpenAI developed an AI system deployed by Pro-New Patriotic Party bot network, which harmed John Mahama , National Democratic Congress , Ghanaian electorate , Democracy と Electoral integrity.
関与が疑われるAIシステム: ChatGPT と X (Twitter)
インシデントのステータス
Risk Subdomain
A further 23 subdomains create an accessible and understandable classification of hazards and harms associated with AI
4.1. Disinformation, surveillance, and influence at scale
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.
- Malicious Actors & Misuse
Entity
Which, if any, entity is presented as the main cause of the risk
Human
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
Intentional
インシデントレポート
レポートタイムライン
ガーナの大統領選挙が12月7日に迫る中、研究者らはX上で、ChatGPTを使用して現職政党である新愛国党（NPP）に有利な投稿をする171のボットアカウントのネットワークを発見した。
AIが作成したボットプロフィールは、ガーナの大統領マハムドゥ・バウミアを宣伝している。 X/@joyce_j5
ニュースメディアの正確性と信頼性を評価するツールを提供するウェブサイトNewsGuardが実施した新しい調査によると、ボットアカウントは一貫してNPP候補者のマハムドゥ・バウミア氏と…
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
よく似たインシデント
Did our AI mess up? Flag the unrelated incidents
よく似たインシデント
Did our AI mess up? Flag the unrelated incidents