Luca Sambucci

Esperto di AI Security, Consulente dell'Unione Europea

Articoli dell'autore



<img width="768" height="432" src="https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/Intelligenza-artificiale-e-manipolazione-emotiva-768x432.jpg" class="card-responsive-img fit-img" alt="Cyber attacchi autonomi: come l’IA sta cambiando il panorama della sicurezza" decoding="async" srcset="https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/Intelligenza-artificiale-e-manipolazione-emotiva-768x432.jpg 768w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/Intelligenza-artificiale-e-manipolazione-emotiva-1280x720.jpg 1280w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/Intelligenza-artificiale-e-manipolazione-emotiva-156x88.jpg 156w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/Intelligenza-artificiale-e-manipolazione-emotiva-1536x864.jpg 1536w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/Intelligenza-artificiale-e-manipolazione-emotiva-640x360.jpg 640w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/Intelligenza-artificiale-e-manipolazione-emotiva-1024x576.jpg 1024w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/Intelligenza-artificiale-e-manipolazione-emotiva-lq.jpg 30w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/Intelligenza-artificiale-e-manipolazione-emotiva.jpg 1920w" sizes="(max-width: 768px) 100vw, 768px" data-eio="l" />

nuove minacce
Manipolazione emotiva: la tecnica che mette a nudo le vulnerabilità dei LLM
22 Ott 2024
di Luca Sambucci
Condividi il post
Condividi



<img width="768" height="431" src="https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/AI-e-difesa-alleato-o-cavallo-di-troia-768x431.jpg" class="card-responsive-img fit-img" alt="AI e difesa alleato o cavallo di troia" decoding="async" srcset="https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/AI-e-difesa-alleato-o-cavallo-di-troia-768x431.jpg 768w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/AI-e-difesa-alleato-o-cavallo-di-troia-1280x718.jpg 1280w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/AI-e-difesa-alleato-o-cavallo-di-troia-156x88.jpg 156w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/AI-e-difesa-alleato-o-cavallo-di-troia-640x360.jpg 640w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/AI-e-difesa-alleato-o-cavallo-di-troia-1024x574.jpg 1024w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/AI-e-difesa-alleato-o-cavallo-di-troia-lq.jpg 30w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/AI-e-difesa-alleato-o-cavallo-di-troia.jpg 1379w" sizes="(max-width: 768px) 100vw, 768px" data-eio="l" />

la riflessione
L’AI per la difesa nazionale: alleata o cavallo di Troia? L'esperimento nel Regno Unito
15 Ott 2024
di Luca Sambucci
Condividi il post
Condividi



<img width="768" height="432" src="https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/AI-e-deepfake-nuove-frontiere-del-phishing-768x432.jpg" class="card-responsive-img fit-img" alt="AI e deepfake nuove frontiere del phishing" decoding="async" srcset="https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/AI-e-deepfake-nuove-frontiere-del-phishing-768x432.jpg 768w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/AI-e-deepfake-nuove-frontiere-del-phishing-1280x720.jpg 1280w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/AI-e-deepfake-nuove-frontiere-del-phishing-156x88.jpg 156w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/AI-e-deepfake-nuove-frontiere-del-phishing-1536x864.jpg 1536w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/AI-e-deepfake-nuove-frontiere-del-phishing-640x360.jpg 640w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/AI-e-deepfake-nuove-frontiere-del-phishing-1024x576.jpg 1024w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/AI-e-deepfake-nuove-frontiere-del-phishing-lq.jpg 30w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/AI-e-deepfake-nuove-frontiere-del-phishing.jpg 1920w" sizes="(max-width: 768px) 100vw, 768px" data-eio="l" />

sicurezza informatica
AI e Deepfake, nuove frontiere del phishing: come difendersi
08 Ott 2024
di Luca Sambucci
Condividi il post
Condividi



<img width="768" height="432" src="https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/Vulnerabilita-memoria-a-lungo-termine-ChatGPT-768x432.jpg" class="card-responsive-img fit-img" alt="Vulnerabilità memoria a lungo termine ChatGPT" decoding="async" srcset="https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/Vulnerabilita-memoria-a-lungo-termine-ChatGPT-768x432.jpg 768w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/Vulnerabilita-memoria-a-lungo-termine-ChatGPT-1280x720.jpg 1280w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/Vulnerabilita-memoria-a-lungo-termine-ChatGPT-156x88.jpg 156w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/Vulnerabilita-memoria-a-lungo-termine-ChatGPT-1536x864.jpg 1536w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/Vulnerabilita-memoria-a-lungo-termine-ChatGPT-640x360.jpg 640w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/Vulnerabilita-memoria-a-lungo-termine-ChatGPT-1024x576.jpg 1024w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/Vulnerabilita-memoria-a-lungo-termine-ChatGPT-lq.jpg 30w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/10/Vulnerabilita-memoria-a-lungo-termine-ChatGPT.jpg 1920w" sizes="(max-width: 768px) 100vw, 768px" data-eio="l" />

l'analisi tecnica
AI e falsi ricordi: il problema della sicurezza nella memoria a lungo termine di ChatGPT
01 Ott 2024
di Luca Sambucci
Condividi il post
Condividi



<img width="768" height="432" src="https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/09/Slack-AI-vulnerabilita-dei-sistemi-AI-768x432.jpg" class="card-responsive-img fit-img" alt="AI generativa e il rischio Lethal trifecta: i tre fattori in gioco" decoding="async" srcset="https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/09/Slack-AI-vulnerabilita-dei-sistemi-AI-768x432.jpg 768w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/09/Slack-AI-vulnerabilita-dei-sistemi-AI-1280x720.jpg 1280w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/09/Slack-AI-vulnerabilita-dei-sistemi-AI-156x88.jpg 156w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/09/Slack-AI-vulnerabilita-dei-sistemi-AI-1536x864.jpg 1536w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/09/Slack-AI-vulnerabilita-dei-sistemi-AI-640x360.jpg 640w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/09/Slack-AI-vulnerabilita-dei-sistemi-AI-1024x576.jpg 1024w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/09/Slack-AI-vulnerabilita-dei-sistemi-AI-lq.jpg 30w, https://dnewpydm90vfx.cloudfront.net/wp-content/uploads/2024/09/Slack-AI-vulnerabilita-dei-sistemi-AI.jpg 1920w" sizes="(max-width: 768px) 100vw, 768px" data-eio="l" />

l'analisi tecnica
Prompt injection e vulnerabilità dei sistemi AI: cosa impariamo dal caso di Slack AI
24 Set 2024
di Luca Sambucci
Condividi il post
Condividi