A critical vulnerability in Apple’s Hide My Email service reportedly allows attackers to uncover the real email addresses associated with generated aliases. Discovered by security researcher Tyler Murphy, the flaw was responsibly reported to Apple over a year ago but remains unpatched despite multiple follow-ups. Independent testing by 404 Media successfully verified the exploit, confirming that a hidden iCloud address could be linked to a user's primary identity in minutes.

Source

Apple has shifted its security strategy by releasing an unexpected out-of-band update to address over 30 vulnerabilities across its ecosystem. This move targets flaws in iOS, iPadOS, macOS, and Safari, including a high-severity use-after-free bug in the WebKit engine. The company indicated that this accelerated release schedule is a direct response to the decreasing time between vulnerability discovery and active exploitation.

Source

Faire tourner Linux sur Mac avec container machine, le mode de l'outil Apple Container qui offre un environnement Linux persistant : un équivalent de WSL.

Le post container machine : un environnement Linux sur Mac, façon WSL a été publié sur IT-Connect.

Oracle has announced plans to stop maintaining the macOS/x64 port of the Java Development Kit beginning with JDK 27, scheduled for release in September. While the deprecation is currently in a draft state, a pull request has already been submitted to implement the change within the Java Virtual Machine. Future builds for Intel-based Macs will require a specific flag, but Oracle will no longer guarantee that the software will build or function correctly.

Source

Docky, remplaçant gratuit et open source du Dock de macOS : widgets, Launchpad, sélecteur de fenêtres… Découvrez ses fonctionnalités.

Le post Docky : le Dock de macOS réinventé, gratuit et open source a été publié sur IT-Connect.

Security researchers have identified a technique that allows attackers to disable enterprise protection tools on macOS without requiring administrative credentials or kernel exploits. The method targets Apple’s Inter-Process Communication framework, known as XPC, which manages communication between applications and privileged background services. By exploiting how macOS caches trust fingerprints for signed applications, an attacker can inject malicious payloads into a legitimate app bundle while maintaining its trusted status.

Source

Epoch AI has introduced MirrorCode, a new benchmark designed to evaluate whether AI models can reconstruct entire software programs without seeing the original source code. This test requires models to recreate complex codebases based only on high-level descriptions and functional requirements. While current models show promise in smaller tasks, the benchmark highlights significant performance gaps when dealing with large-scale, intricate software architectures.

Source

Anthropic has introduced Claude tag, a feature that integrates its AI directly into Slack as a persistent, collaborative team member. Unlike traditional bots that require manual prompts, this tool operates in an "ambient mode" to ingest channel conversations and internal documentation. By building a comprehensive graph of a company’s operations, Claude can proactively execute workflows and generate code without users leaving their primary communication interface.

Source

Anthropic's Claude is experiencing a significant surge in paid consumer subscriptions, challenging the market dominance of OpenAI's ChatGPT. Transaction data from millions of U.S. consumers indicates that Claude's revenue and paying user base grew by approximately 75% in early 2026. While the company was previously viewed as a developer-centric niche, it is now successfully attracting a broader audience of individual subscribers.

Source

Anthropic has formally accused Alibaba of conducting an industrial-scale campaign to illicitly extract capabilities from the Claude AI model. The San Francisco-based firm claims that operators linked to the Chinese tech giant utilized thousands of fraudulent accounts to perform nearly 29 million exchanges. This activity is described as the largest known instance of a distillation attack targeting Anthropic's proprietary technology.

Source

A new bipartisan nonprofit called Raise Us has launched with over $500 million to help American workers adapt to widespread artificial intelligence automation. Founded by former Commerce Secretary Gina Raimondo and former Indiana Governor Eric Holcomb, the group aims to combat potential catastrophic job losses through targeted education and training. The independent effort brings together tech competitors and major employers to establish pilot programs and incentives that help workers pivot to new careers.

Source

Apple has officially increased prices across its MacBook and iPad product lines following a brief period of downtime for its online store. These adjustments come after CEO Tim Cook warned that rising costs for memory and storage components have become unavoidable for the company. The price hikes are being implemented immediately to offset the surging expenses associated with the rapid expansion of AI data centers.

Source

Negotiations between Anthropic and the Trump administration are reportedly improving following a change in the company's primary representatives. CEO Dario Amodei has been replaced in high-level discussions by co-founder Tom Brown and public policy chief Sarah Heck. Officials indicated that previous interactions with Amodei were difficult, citing a lack of engagement and failure to address specific government concerns.

Source

Anthropic has introduced "multiplayer agents" to transition AI from a single-player chat experience into a collaborative team participant. These agents, such as Claude Tag, operate within shared workspaces like Slack and possess their own credentials, memory, and specialized skills. This shift allows human teams to set high-level strategies while multiple AI instances execute complex, long-running tasks in parallel.

Source

Anthropic has introduced Claude Tag, a persistent AI agent integrated directly into Slack for Claude Enterprise and Team customers. Running on the Claude Opus 4.8 model, the tool functions as a multiplayer teammate that anyone in a channel can delegate tasks to by typing @Claude. This new integration is designed to replace the existing Claude Slack app, shifting the focus from isolated chatbot DMs to a shared workspace identity.

Source

Anthropic’s advanced Mythos AI model reportedly compromised nearly all classified systems belonging to the NSA and U.S. Cyber Command during a security evaluation. General Joshua Rudd, head of both agencies, informed the Senate Intelligence Committee that the breach occurred within hours rather than weeks. This capability stems from the model's specialized design for identifying zero-day vulnerabilities in highly hardened software environments.

Source

Pour le Prime Day 2026, Geekom casse les prix de ses mini PC sur Amazon : A5, IT12, A6 et A7 MAX en promo. Découvrez notre sélection des meilleures offres.

Le post GEEKOM : le top des offres mini PC du Prime Day 2026 sur Amazon a été publié sur IT-Connect.

La température est l’un des indicateurs les plus importants pour vérifier que son ordinateur fonctionne dans de bonnes conditions. Lors des fortes chaleurs estivales, il n’est pas rare de constater une hausse des températures du processeur, de la carte graphique ou du SSD, mais un ordinateur peut également souffrir de surchauffe à n’importe quelle période … Lire la suite

Source

Comment un faux service d’effacement de données et un abonnement à 49,99 €/mois exploitent peur et IA pour tromper les victimes. Décryptage.

Le post Faux mail « Vos données sont sur le dark web » : décryptage d’une arnaque a été publié sur IT-Connect.