OpenAI Unveils $1 Billion Initiative to Safeguard Essential Services
OpenAI has launched Daybreak for Frontline Defenders, a $1 billion initiative aimed at enhancing access to frontier cyber AI, training, and support for essential services.
Breaking
Topic
Threats, privacy, and enterprise security.
OpenAI has launched Daybreak for Frontline Defenders, a $1 billion initiative aimed at enhancing access to frontier cyber AI, training, and support for essential services.
OpenAI has announced that GPT-6 Astra is its most advanced model, achieving a Critical level of cybersecurity capability as per its Preparedness Framework.
Reolink has launched the OMVI 2i Ultra security camera, featuring dual cameras for wide and focused views. It includes upgraded AI for custom scenarios and a $199.99 Home Hub 2 for local storage and video search. The camera aims to enhance home security, especially for families with pets or children
Broadcom has issued security updates for critical vulnerabilities in VMware Workstation and Fusion, including CVE-2026-59346, which allows local attackers to execute arbitrary code due to an integer-overflow flaw.
JetBrains has advised Cadence users to revoke and rotate their credentials following a security breach caused by a vulnerability in TeamCity. The breach allowed attackers to access JetBrains' environment.
A new unpatched vulnerability in Magento Open Source and Adobe Commerce, named StyleSmuggler, is being exploited by attackers to run malicious code on online stores' servers without authentication, according to Sansec.
Elastic Security Labs has identified four new programs linked to the REVSTEALER malware that disable Windows Update and Microsoft Defender to facilitate cryptocurrency mining on infected machines.
Attackers are exploiting MikroTik routers via their internet-exposed SSH service, gaining full control without authentication, as warned by CERT Polska on September 5. The attacks have been occurring since at least September 2.
Trezor announced that a data breach at its shipping provider ShipMonk has exposed the information of 67,000 U.S. customers. The compromised data includes names, email addresses, phone numbers, shipping addresses, and order numbers from November 2019 to August 2021, but the security of Trezor's hardw
CISA has added seven security vulnerabilities to its Known Exploited Vulnerabilities catalog, highlighting the risks of reverse shells and crypto miners targeting these flaws.
A security researcher known as Chaotic Eclipse has released a zero-day proof-of-concept named FalconFlank, which demonstrates a privilege escalation vulnerability in CrowdStrike Falcon. The exploit targets the remediation of malicious macros in the CrowdStrike Falcon Sensor.
A member of Serbia's student protest movement had their iPhone infected with NSO Group's Pegasus spyware via a zero-click exploit, as confirmed by Citizen Lab and SHARE Foundation.
GitGuardian researchers discovered that the Shai-Hulud infostealer worm has expanded its scanning capabilities to 469 credential locations, up from 189. This significant increase highlights the evolving threat posed by attackers in developer environments and cloud configurations.
Threat actors are exploiting the Node.js runtime to deliver malware in targeted attacks against various sectors, including government and technology, as reported by Symantec.
A phishing campaign initially linked to Canada has expanded to 46 countries, with the US being the primary target, accounting for 45% of the activity. Research by ANY.RUN identified 601 cases connected to this operation.
Thomson Reuters reported a data breach involving its C-Track court management platform, affecting courts in 11 U.S. states, the U.S. Virgin Islands, and Ontario, Canada. The breach was discovered on June 30, 2026, and may have exposed sensitive information.
Researchers have revealed BraZetsu, a sophisticated Python-based malware framework that transforms compromised Windows hosts into assets for an underground marketplace, aiding Initial Access Brokers.
The latest edition of ThreatsDay highlights the normalization of phishing attacks, where attackers exploit trust through fake login pages and shared files. The report emphasizes the ease with which these attacks can occur, often requiring just a simple click or a minor mistake in a web address.
PostgreSQL has released updates to fix a 12-year-old security flaw that allows accounts with the REPLICATION attribute to execute arbitrary code as the database server's operating system user. The flaw, tracked as CVE-2026-6471, has existed since the introduction of logical decoding in 2014.
Google has released a security update for Chrome, addressing 12 vulnerabilities, including a high-severity zero-day vulnerability in the V8 engine, tracked as CVE-2026-85046.
Microsoft has issued a warning about a high-volume phishing campaign that employs invisible Unicode characters to evade email filters. This tactic allows attackers to manipulate financial lure words, making them harder for filters to detect.
Threat actors are exploiting newly disclosed vulnerabilities in PaperCut to steal credentials from educational institutions in the U.S. and Europe. The Arctic Wolf Adversary Research Team reported that these flaws enable command execution and reconnaissance.
Plex has urged users to update to the latest versions of Plex Media Server and Plex Desktop after patching multiple undisclosed security flaws. The company has requested CVE identifiers for these issues.
Threat actors are exploiting critical vulnerabilities in WordPress plugins Super Forms and Elementor Pro, with over 440,000 exploit attempts reported. The vulnerabilities allow unauthenticated file uploads, posing significant security risks.