Is superintelligence necessarily moral?

Numerous authors have expressed concern that advanced artificial intelligence (AI) poses an existential risk to humanity. These authors argue that we might build AI which is vastly intellectually superior to humans (a 'superintelligence'), and which optimizes for goals that strike us as morally bad,...

Descripción completa

Detalles Bibliográficos
Publicado en:Analysis Vol. 84; no. 4; pp. 730 - 739
Autor principal: Dung, Leonard
Formato: Artículo
Publicado: Oxford University Press / USA Oct2024
Materias:
Acceso en línea:Ver este registro en EBSCOhost
fields @attributes:
  recordID: 1
pdfLink:
plink: https://search.ebscohost.com/login.aspx?direct=true&db=hlh&AN=183906880&site=ehost-live
header:
  @attributes:
    shortDbName: hlh
    uiTerm: 183906880
    longDbName: Humanities International Complete
    uiTag: AN
  controlInfo:
    bkinfo:
    jinfo:
      jid:
        00032638
        7OX
      jtl: Analysis
      issn: 00032638
      maglogo: N
    pubinfo:
      dt: Oct2024
      vid: 84
      iid: 4
      pid: 622
      pub: Oxford University Press / USA
    artinfo:
      ui:
        183906880
        10.1093/analys/anae033
      ppf: 730
      ppct: 9
      formats:
        fmt:
          – @attributes:
              type: T
          – @attributes:
              type: P
              size: 292KB
      tig:
        atl: Is superintelligence necessarily moral?
      aug:
        au: Dung, Leonard
        affil: University Erlangen-Nürnberg   Germany
      su:
        Artificial intelligence
        Realism
        Externalism (Philosophy of mind)
        Moral reasoning
        Humanity
      sug:
        subj:
          Artificial intelligence
          Realism
          Externalism (Philosophy of mind)
          Moral reasoning
          Humanity
      keyword:
        AI alignment
        existential risk
        moral motivation
        orthogonality thesis
        reward hacking
        superintelligence
      ab: Numerous authors have expressed concern that advanced artificial intelligence (AI) poses an existential risk to humanity. These authors argue that we might build AI which is vastly intellectually superior to humans (a 'superintelligence'), and which optimizes for goals that strike us as morally bad, or even irrational. Thus this argument assumes that a superintelligence might have morally bad goals. However, according to some views, a superintelligence necessarily has morally adequate goals. This might be the case either because abilities for moral reasoning and intelligence mutually depend on each other, or because moral realism and moral internalism are true. I argue that the former argument misconstrues the view that intelligence and goals are independent, and that the latter argument misunderstands the implications of moral internalism. Moreover, the current state of AI research provides additional reasons to think that a superintelligence could have bad goals.
      pubtype: Academic Journal
      doctype: Article
      src: R
    language: English
    refInfo:
    copyright:
      @attributes:
        flag: Y
      custom: © 2019 The Analysis Trust.
      item: Analysis
      holder: Oxford University Press / USA
      dt:
        @attributes:
          year: 2024
    holdings:
      @attributes:
        islocal: N