benchmark

approfondimento

Red teaming, difesa a strati: le misure per ridurre il rischio cyber degli LLM

13 Mar 2026

<img width="768" height="512" src="https://img.ai4business.it/wp-content/uploads/2025/09/19104013/word-image-47984-1-768x512.jpg" class="card-responsive-img fit-img" alt="SWE-bench Verified" decoding="async" srcset="https://img.ai4business.it/wp-content/uploads/2025/09/19104013/word-image-47984-1-768x512.jpg 768w, https://img.ai4business.it/wp-content/uploads/2025/09/19104013/word-image-47984-1-1080x720.jpg 1080w, https://img.ai4business.it/wp-content/uploads/2025/09/19104013/word-image-47984-1-132x88.jpg 132w, https://img.ai4business.it/wp-content/uploads/2025/09/19104013/word-image-47984-1-1024x683.jpg 1024w, https://img.ai4business.it/wp-content/uploads/2025/09/19104013/word-image-47984-1-lq.jpg 30w, https://img.ai4business.it/wp-content/uploads/2025/09/19104013/word-image-47984-1.jpg 1536w" sizes="(max-width: 768px) 100vw, 768px" data-eio="l" />

tecnologia
SWE-bench Verified non misura più le capacità di coding di frontiera. Quale alternativa?
25 Feb 2026
di Fabio Lalli
Condividi il post
Condividi



<img width="768" height="512" src="https://img.ai4business.it/wp-content/uploads/2026/02/06124458/image-40-768x512.jpg" class="card-responsive-img fit-img" alt="GPT-5.3 Codex" decoding="async" srcset="https://img.ai4business.it/wp-content/uploads/2026/02/06124458/image-40-768x512.jpg 768w, https://img.ai4business.it/wp-content/uploads/2026/02/06124458/image-40-1080x720.jpg 1080w, https://img.ai4business.it/wp-content/uploads/2026/02/06124458/image-40-132x88.jpg 132w, https://img.ai4business.it/wp-content/uploads/2026/02/06124458/image-40-1024x683.jpg 1024w, https://img.ai4business.it/wp-content/uploads/2026/02/06124458/image-40-lq.jpg 30w, https://img.ai4business.it/wp-content/uploads/2026/02/06124458/image-40.jpg 1536w" sizes="(max-width: 768px) 100vw, 768px" data-eio="l" />

ai coding
GPT-5.3-Codex: l’agente AI che lavora sul computer come un professionista
06 Feb 2026
di Pierluigi Sandonnini
Condividi il post
Condividi



<img width="768" height="768" src="https://img.ai4business.it/wp-content/uploads/2025/11/25131010/image-40-768x768.png" class="card-responsive-img fit-img" alt="Claude Opus 4.5" decoding="async" loading="lazy" srcset="https://img.ai4business.it/wp-content/uploads/2025/11/25131010/image-40-768x768.png 768w, https://img.ai4business.it/wp-content/uploads/2025/11/25131010/image-40-720x720.png 720w, https://img.ai4business.it/wp-content/uploads/2025/11/25131010/image-40-88x88.png 88w, https://img.ai4business.it/wp-content/uploads/2025/11/25131010/image-40-120x120.png 120w, https://img.ai4business.it/wp-content/uploads/2025/11/25131010/image-40-lq.png 30w, https://img.ai4business.it/wp-content/uploads/2025/11/25131010/image-40.png 1000w" sizes="auto, (max-width: 768px) 100vw, 768px" data-eio="l" />

approfondimento
Claude per i professionisti e le aziende: il punto di forza sta nell'ecosistema
26 Gen 2026
di Paolo Dalprato
Condividi il post
Condividi



<img width="768" height="512" src="https://img.ai4business.it/wp-content/uploads/2026/01/22105741/image-22-5-768x512.jpg" class="card-responsive-img fit-img" alt="valutazione agenti AI" decoding="async" loading="lazy" srcset="https://img.ai4business.it/wp-content/uploads/2026/01/22105741/image-22-5-768x512.jpg 768w, https://img.ai4business.it/wp-content/uploads/2026/01/22105741/image-22-5-1080x720.jpg 1080w, https://img.ai4business.it/wp-content/uploads/2026/01/22105741/image-22-5-132x88.jpg 132w, https://img.ai4business.it/wp-content/uploads/2026/01/22105741/image-22-5-1024x683.jpg 1024w, https://img.ai4business.it/wp-content/uploads/2026/01/22105741/image-22-5-lq.jpg 30w, https://img.ai4business.it/wp-content/uploads/2026/01/22105741/image-22-5.jpg 1536w" sizes="auto, (max-width: 768px) 100vw, 768px" data-eio="l" />

guida
Valutazione degli agenti AI in azienda: framework, architettura e gestione operativa
22 Gen 2026
di Fabio Lalli
Condividi il post
Condividi



<img width="493" height="488" src="https://img.ai4business.it/wp-content/uploads/2025/12/15111305/image-31.png" class="card-responsive-img fit-img" alt="GPT-5.2 analisi" decoding="async" loading="lazy" srcset="https://img.ai4business.it/wp-content/uploads/2025/12/15111305/image-31.png 493w, https://img.ai4business.it/wp-content/uploads/2025/12/15111305/image-31-89x88.png 89w, https://img.ai4business.it/wp-content/uploads/2025/12/15111305/image-31-120x120.png 120w, https://img.ai4business.it/wp-content/uploads/2025/12/15111305/image-31-lq.png 30w" sizes="auto, (max-width: 493px) 100vw, 493px" data-eio="l" />

approfondimento
Dentro GPT-5.2: cosa dicono davvero i benchmark
15 Dic 2025
di Giovanni Masi
Condividi il post
Condividi



<img width="768" height="427" src="https://img.ai4business.it/wp-content/uploads/2025/12/12100659/image-14-1-768x427.jpg" class="card-responsive-img fit-img" alt="GPT-5.2" decoding="async" loading="lazy" srcset="https://img.ai4business.it/wp-content/uploads/2025/12/12100659/image-14-1-768x427.jpg 768w, https://img.ai4business.it/wp-content/uploads/2025/12/12100659/image-14-1-156x88.jpg 156w, https://img.ai4business.it/wp-content/uploads/2025/12/12100659/image-14-1-1024x569.jpg 1024w, https://img.ai4business.it/wp-content/uploads/2025/12/12100659/image-14-1-lq.jpg 30w, https://img.ai4business.it/wp-content/uploads/2025/12/12100659/image-14-1.jpg 1237w" sizes="auto, (max-width: 768px) 100vw, 768px" data-eio="l" />

ai generativa
GPT-5.2, il nuovo standard dell’AI professionale di OpenAI
12 Dic 2025
di Pierluigi Sandonnini
Condividi il post
Condividi



<img width="657" height="408" src="https://img.ai4business.it/wp-content/uploads/2025/11/19093535/image-18.png" class="card-responsive-img fit-img" alt="Google Gemini 3 prova" decoding="async" loading="lazy" srcset="https://img.ai4business.it/wp-content/uploads/2025/11/19093535/image-18.png 657w, https://img.ai4business.it/wp-content/uploads/2025/11/19093535/image-18-142x88.png 142w, https://img.ai4business.it/wp-content/uploads/2025/11/19093535/image-18-lq.png 30w" sizes="auto, (max-width: 657px) 100vw, 657px" data-eio="l" />

tecnologia
Gemini 3 alla prova: cosa dicono veramente i test sul nuovo modello di Google
26 Nov 2025
di Giovanni Masi
Condividi il post
Condividi



<img width="768" height="379" src="https://img.ai4business.it/wp-content/uploads/2025/11/10133756/word-image-49604-3-768x379.png" class="card-responsive-img fit-img" alt="OpenAI evals" decoding="async" loading="lazy" srcset="https://img.ai4business.it/wp-content/uploads/2025/11/10133756/word-image-49604-3-768x379.png 768w, https://img.ai4business.it/wp-content/uploads/2025/11/10133756/word-image-49604-3-156x77.png 156w, https://img.ai4business.it/wp-content/uploads/2025/11/10133756/word-image-49604-3-1024x505.png 1024w, https://img.ai4business.it/wp-content/uploads/2025/11/10133756/word-image-49604-3-lq.png 30w, https://img.ai4business.it/wp-content/uploads/2025/11/10133756/word-image-49604-3.png 1028w" sizes="auto, (max-width: 768px) 100vw, 768px" data-eio="l" />

tecnologia
Come gli evals stanno guidando l'introduzione dell’AI nelle imprese
20 Nov 2025
di Alessandra Castelli
Condividi il post
Condividi