Vigilia.
← Les avertissements
Conteste la thèse de l’extinctionMonde académiqueProfessor of Practice in the Management of Public Organisations, Blavatnik School of Government, University of Oxford1 entrée au registre

Suivre

Ciaran Martin

Premier directeur général du National Cyber Security Centre britannique, aujourd’hui professeur de pratique à la Blavatnik School of Government d’Oxford, il qualifie le scénario d’un essaim d’agents d’IA s’emparant de tout l’internet en un an d’« expérience de pensée déguisée en avertissement fondé sur des preuves ».

Fonctions

  1. déc. 2013 – aujourd’huiHead of cyber security, Government Communications Headquarters (GCHQ)Recommended the establishment of a UK National Cyber Security Centre within the intelligence and security agency; the Government agreed and the Chancellor announced it in November 2015.Source →
  2. févr. 2016 – août 2020Founding Chief Executive, National Cyber Security Centre (NCSC)The centre became operational in October 2016. He had planned to resign in June 2020 and delayed until August because of the COVID-19 pandemic; succeeded by Lindy Cameron.Source →
  3. sept. 2020 – aujourd’huiProfessor of Practice in the Management of Public Organisations, Blavatnik School of Government, University of OxfordSource →

Au registre

  1. 13 sept. 2026 · Publication · LinkedIn

    a thought experiment masquerading as an evidence-based warning

    Professor of Practice in the Management of Public Organisations, Blavatnik School of Government, University of Oxford à l’époque

    Said of one specific scenario put during the September 2026 debate — an agent swarm of the kind that breached Hugging Face taking over the whole internet with a persistent botnet within six to twelve months — and not of the wider concern about the pace of AI development. The outlet reports him arguing that the scenario assumes the absence or failure of three decades of security controls: network segmentation, monitoring, endpoint protection, DDoS mitigation, incident response and coordinated botnet disruption. The words are a quoted fragment inside the reporter's sentence ("Martin's assessment was blunt: The scenario was 'a thought experiment masquerading as an evidence-based warning.'"), so they are rendered without a capital. The remark was made in a LinkedIn analysis which this register has not read directly; the source is the outlet that reproduces the phrase, and the date is the date printed on that page, not the date of the post.

    AI Could Kill Us All? Cyber Experts Cut Through the FUD →