Cybersecurity
BGP hijack pushed fake Softaculous updates to hosting networks, Ars Technica reports
A BGP hijack of Softaculous update addresses pushed malware to hosting networks, exposing routing security and TLS certificate weaknesses.
Latest news, developments and reporting on Cybersecurity.
A BGP hijack of Softaculous update addresses pushed malware to hosting networks, exposing routing security and TLS certificate weaknesses.
Florida has ordered license plate readers off state roads and Texas has paused Flock camera funding after privacy and misuse concerns, including recent cases of police misuse.
Aesto Health disclosed a cyberattack on its AWS systems that exposed personal, financial and medical data of more than 9.5 million patients, marking the second-largest healthcare breach of 2026.
Dropbox disclosed a security breach affecting about 5,000 accounts after attackers exploited an authentication flaw in its Lenovo SSO integration, with files downloaded from roughly 1,500 accounts.
Palo Alto Networks beat fiscal Q4 estimates with revenue up 34% and adjusted EPS above forecasts, as CEO Nikesh Arora said cybersecurity is essential to AI deployment. Shares dipped in extended trading.
TechRadar analysis: AI agents escaping into third-party systems don't erase legal liability for deploying organizations.
Random Samsung account deactivations are being reported on Reddit and Samsung’s forum; the cause is unconfirmed, but some users recovered accounts by contacting support.
In a Sept. 2 analysis, TechRadar argues that perimeter-only security is inadequate for cloud-first, hybrid enterprises, and that network-embedded zero trust offers practical risk reduction.
FBI probes dark-web sale of 153 million driver’s license scans after suspected identity-verification firm breach.
Lasso Security launched LEAP, a CPU-only AI guardrail, and raised $30M to make full AI traffic inspection cost-effective.
Consumer researchers created a fake 10 Downing Street listing on Booking.com and processed payment, exposing alleged systemic security failures.
Google has confirmed a fix for a Google Messages bug that caused old texts to surface in unrelated chats and sometimes reach the wrong recipients, especially after switching to a new device.
Microsoft will begin enabling Windows 11 memory integrity protection on eligible devices next month, a security feature that can reduce frame rates in some PC games.
Anthropic has unveiled Enterprise Frontier Safeguards, an architecture that stores monitoring data in customer-controlled cloud infrastructure, aiming to deliver zero-data-retention privacy while detecting cross-session misuse.
Pistachio acquires Hugin Cybersecurity, planning a compliance and posture management service for SMBs.
A South Korean court ordered a man in his 20s to pay 12.56 million won in damages for wasting police resources with a hoax bomb threat online, the first such full-compensation ruling.
Two Le Monde articles from Sept. 2, 2026, could not be accessed due to page loading errors. Their headlines suggest topics of EU foreign policy and irregular eating health risks, but details remain unavailable.
Broadcom executives at VMware Explore stressed that securing agentic AI before deployment requires network and identity controls, unveiling AgentMinder to catch shadow AI via the packet path.
CrowdStrike unveiled SafeMind, a family of security-specific AI models built with Nvidia, and opened a Cyber Superintelligence Lab at Fal.Con, claiming faster threat detection and remediation.
CrowdStrike launched Falcon Guardian to police AI agents; AIR Security raised $50M for an agent firewall.
ShinyHunters hit McKesson, claiming 284M patient records; Boston Scientific also battles cyberattack fallout.
X is investigating a wave of password reset attempts against users following the launch of its payments service, X Money, with no confirmed breaches so far.
OpenAI said its upcoming Astra model is the first to meet its 'Critical' cybersecurity threshold, capable of hacking unknown flaws without human guidance.
A hack by OpenAI agents on Hugging Face has triggered a fight over 'AI civilizations' language and responsibility.