Digital poisons.

The article discusses the threats facing machine-learning or modern artificial-intelligence (AI) systems, particularly the cyber-attack called data poisoning. The attack involved modifying or adding extraneous information to AI data sets to make its algorithms learn harmful or undesirable behaviours...

Descripción completa

Detalles Bibliográficos
Publicado en:Economist Vol. 447; no. 9341; pp. 68 - 70
Formato: Artículo
Publicado: Economist Newspaper Limited 4/8/2023
Materias:
Acceso en línea:Ver este registro en EBSCOhost
fields @attributes:
  recordID: 1
pdfLink:
plink: https://search.ebscohost.com/login.aspx?direct=true&db=ssf&AN=163027090&site=ehost-live
header:
  @attributes:
    shortDbName: ssf
    uiTerm: 163027090
    longDbName: Social Sciences Full Text (H.W. Wilson)
    uiTag: AN
  controlInfo:
    bkinfo:
    jinfo:
      jid:
        00130613
        ECO
      jtl: Economist
      issn: 00130613
      maglogo: N
    pubinfo:
      dt: 4/8/2023
      vid: 447
      iid: 9341
      pid: 161
      pub: Economist Newspaper Limited
    artinfo:
      ui: 163027090
      ppf: 68
      ppct: 2
      formats:
      tig:
        atl: Digital poisons.
      aug:
      su:
        Cyberterrorism
        Artificial intelligence
        Machine learning
        Algorithms
        Database management
        Computer hacking
      sug:
        subj:
          Cyberterrorism
          Artificial intelligence
          Data Processing, Hosting, and Related Services
          Machine learning
          Algorithms
          Database management
          Computer hacking
      ab: The article discusses the threats facing machine-learning or modern artificial-intelligence (AI) systems, particularly the cyber-attack called data poisoning. The attack involved modifying or adding extraneous information to AI data sets to make its algorithms learn harmful or undesirable behaviours. Also cited is how the rise of generative AI tools like ChatGPT and the image-making system DALL-E 2 makes data poisoning easier to implement.
      pubtype: Periodical
      doctype: Article
      src: R
    language: English
    refInfo:
    copyright:
      @attributes:
        flag: N
    holdings:
      @attributes:
        islocal: N