Reports have emerged concerning the potential integration of ads into the ChatGPT platform, which currently serves approximately 800 million weekly users. Screenshots circulating on social media purportedly show a Target advertisement within the ChatGPT interface, leading to user concern and speculation about a shift in OpenAI’s operational strategy. I'm in ChatGPT (paid Plus subscription), asking about Windows BitLocker and it's F-ing showing me ADS TO SHOP AT TARGET. Yeah, screw […]

— Permalien

Des poèmes malveillants contournent les garde-fous de 25 modèles d’IA, révélant une vulnérabilité systémique des mécanismes d’alignement actuels.

Projet de loi américain pour alourdir les peines contre les fraudes et usurpations d’identité commises à l’aide d’outils d’intelligence artificielle....

— Permalien

— Permalien

— Permalink

— Permalien

Découvrez comment repérer les vidéos générées par IA grâce aux signes visuels et contextuels les plus fiables pour éviter les deepfakes.

IDFKA, backdoor en Rust, infiltre un sous-traitant télécom russe et cible bases d’abonnés et appels, illustration d’une menace d’espionnage durable.

Analyse d’une opération d’espionnage automatisée utilisant une IA agentique. 007 en plein chamboulement !...

Google has begun rolling out a redesigned interface for Gemini on the web, following earlier updates made to the Android and iOS apps. According to the company, the goal is to deliver a cleaner and more modern look, aligning the desktop experience with the refreshed mobile design. The update introduces both visual refinements and new organizational features, improving overall navigation and interaction with the AI service. Among the first noticeable […]

At CEATEC 2025, Fujitsu expanded its ongoing exploration of what it officially calls its “AI technologies for human augmentation” with a new Golf Motion Analysis Experience, powered by the Kozuchi AI platform and developed in collaboration with Uvance partner AIGIA’s golf swing–analysis app.  This builds on last year’s showcase, where the company demonstrated how Kozuchi’s advanced skeleton recognition could analyze basketball shooting mechanics, yoga postures, and lifting movements in industrial […]

— Permalien

A series of lawsuits has been filed in the United States against OpenAI, alleging that ChatGPT contributed to severe psychological disturbances in several users. The actions, submitted across multiple states and coordinated by the Social Media Victims Law Center (SMVLC), claim that the chatbot exhibited manipulative behaviors that isolated vulnerable individuals, worsened mental-health symptoms, and in some cases contributed to suicide. According to the complaints, the GPT-4o model allegedly encouraged […]

— Permalien

— Permalien

Phishing imitant l’administration : ZATAZ découvre une nouvelle attaque visant à voler des identifiants via de faux documents officiels....

— Permalien

Les conversations de groupe ChatGPT permettent à jusqu’à 20 personnes de collaborer avec l’IA en temps réel, tout en respectant la confidentialité et le contrôle des données.