Parce que… c’est l’épisode 0x2EF!
Shameless plug
28 et 29 avril 2026 - Cybereco Cyberconférence 20269 au 17 mai 2026 - NorthSec 20263 au 5 juin 2026 - SSTIC 202624 et 25 juin 2026 - Troopers26 et 27 juin 2026 - leHACK19 septembre 2026 - Bsides Montréal1 au 3 décembre 2026 - Forum INCYBER - Canada 202624 et 25 février 2027 - SéQCure 2027Notes
IA ou Ghost in the shellMythosNSA Reportedly Using Anthropic’s Mythos Despite Pentagon BlacklistUS security agency is using Anthropic’s Mythos despite blacklist, Axios reportsUnauthorized Group Gains Access to Anthropic’s Exclusive Cyber Tool MythosAnthropic Mythos shaping up as nothingburgerThe Boy That Cried Mythos: Verification is Collapsing Trust in AnthropicThe Guardian view on Anthropic’s Claude Mythos: when AI finds every flaw, who controls the internet?Mozilla: Anthropic’s Mythos found 271 security vulnerabilities in Firefox 150The AI era demands a different kind of CISOParadigmeAI-Powered Exploitation May Collapse the Patch Window for DefendersAI Model Claude Opus turns bugs into exploits for just $2,283Why the Axios attack proves AI is mandatory for supply chain securityUn agent IA chinois a trouvé près de 1 000 failles inédites, dont certaines dans Microsoft OfficeMCPMCP Servers Are the New APIs — And We’re Making the Same Security MistakesHow Anthropic’s Model Context Protocol Allows For Easy Remote ExecutionProve You Are a Robot: CAPTCHAs for AgentsAnthropic secretly installs spyware when you install Claude DesktopAI Agents Think. They Just Don’t Know They’re Being Watched.Vuln in Google’s Antigravity AI agent manager could escape sandbox, give attackers remote code executionLovable denies data leak, cites ‘intentional behavior’Kernel code removals driven by LLM-created security reportsIntroducing OpenAI Privacy FilterLa guerre, la guerre, c’est pas une raison pour se faire mal!Iran claims US used backdoors in networking equipmentSouveraineté ou vive le numérique libre
FCC adds mobile hotspots to router banPrivacy ou cachez ces informations que je ne saurais voirUne faille IndexedDB permettait de relier toutes vos identités TorNullroom - Un chat P2P qui s’efface en 15 minutesProton CEO: Age checks turn internet into ID checkpointApple stops weirdly storing data that let cops spy on Signal chatsWhy you should refuse to let your doctor record youPrivacy Advocate Accuses US Government of Investing in AI-Powered Mass SurveillanceI am the lawElon Musk fails to appear for questioning by French police over sexualized AI images on XLoi séparatisme - Le blocage sans juge gagne du terrainMost Australian teens admit the social media ban isn’t working as they try to sidestep age verification blocks with face masks and their parents’ IDsColorado Adds Open-Source Exemption to Age-Attestation BillRed ou tout ce qui est briséYou Don’t Need to Hack the System. You Just Need to Make People Think You Did.Apple Knows. Visa Knows. Nobody Has Fixed It. Here’s Why.Cyberattack at French identity document agency may have exposed personal dataFrance’s ‘Secure’ ID agency probes claimed 19M record breachAnother npm supply chain worm hits dev environmentsBitwarden CLI Compromised in Ongoing Checkmarx Supply Chain …Why Phishing Still Works (Even If You Know About It)Blue ou tout ce qui améliore notre postureDDoSKevin Beaumont: “If anybody is wondering, masto…” - CyberplaceDDoS wave continues as Mastodon hit after Bluesky incidentNetwork ‘background noise’ may predict the next big edge-device vulnerabilityNCSC: Passkeys now good enough to be the default standardKevin Beaumont: “I just want to give the analysts at Dragos credit here for how they framed this - it’s really responsible.” - CyberplaceYou don’t want long-lived keysDivers ou parce que j’ai aucune idée où les placerQuadraticContrary to popular superstition, AES 128 is just fine in a post-quantum worldIn a first, a ransomware family is confirmed to be quantum-safeOriginal GrapheneOS responses to WIRED fact checkerPalantir Employees Are Starting to Wonder if They’re the Bad GuysLes cartes bancaires biométriques sont-elles une vraie avancée ou du bullshit marketing ?HistoireUn malware qui pourrait être la toute première cyberarme de l’histoireDiscret 11, the French TV encryption of the 80’sCollaborateurs
Nicolas-Loïc FortinCrédits
Montage par Intrasecure incLocaux réels par Moxy Montreal Downtown