Starlink users appear to have been singled out by a suspicious online advertising operation, according to new research and Reuters reporting that connects the activity to Israeli ad-tech company Discover-Tech. …
Security firm Forever Security reported on Oct. 22, 2026, that it stopped a live attempt in which a plain-text AGENTS.md file in a customer’s Downloads folder steered an AI coding …
The Office of the Privacy Commissioner of Canada said on Oct. 27, 2026, that Commissioner Philippe Dufresne has opened an investigation into Navigate360 related to a cybersecurity incident involving tips …
The Canadian Anti-Fraud Centre (CAFC) said Canadians reported approximately $351 million in fraud losses in the first six months of 2026. The first-half 2026 total should not be treated as …
LiteLLM says its public release pipeline was compromised on October 21, 2026, producing a malicious litellm==1.107.0 release on PyPI and compromised public Docker images. LiteLLM’s incident post says the affected …
Toronto-based 1Password announced 1Password for Agents on October 22, a platform designed to let supported personal AI agents sign in to accounts without seeing or storing the passwords, secrets, or …
Roundcube has released security updates for the 1.6 LTS and 1.7 branches of Roundcube Webmail, moving those branches to 1.6.20 and 1.7.5 with fixes for 20 security issues. The Roundcube …
Several Shopify app vendors have acknowledged unauthorized changes that caused some storefronts to send shoppers from a checkout button to fake payment pages in October 2026. Public reports and vendor …
Railway says an October security incident may have exposed customer environment variables after an unauthorized third party gained access to a subset of its internal systems. Railway first disclosed the …
Rank Math has warned that some Rank Math PRO 3.0.123 packages downloaded or installed through updates during an 11-hour window on Oct. 15 and 16 contained malware, turning what would …
OVHcloud says a compromised developer GitHub account was used to replace official release binaries for its OVHcloud CLI with secret-stealing malware on Oct. 19, 2026. The affected tool is a …
Attackers broke into self-hosted Nextcloud servers through an optional app that adds document editing. The flaw, tracked as CVE-2026-110447, sits in Collabora’s Built-in CODE Server app and lets an attacker …
October 20, 2026 update: ConnectWise Cloud Backup, SaaS Security, and Cloud Management remain suspended following unauthorized access to application identities associated with the former SkyKick service. ConnectWise said it is …
Adobe has released an October security update for Adobe Commerce and Magento Open Source that fixes 11 vulnerabilities, including a maximum-score flaw that can be exploited without authentication to achieve …
BeyondTrust has remediated a critical vulnerability in its Remote Support product after the flaw was exploited as a zero-day in late August, according to the company and Google Cloud’s Mandiant …
Let’s Encrypt has started issuing 64-day TLS certificates in its staging environment for the default classic profile, according to an API announcement on October 14, 2026. TLS certificates are often …
An image of Mia Goth that spread on Instagram this month was AI-generated, according to her representatives. Goth filed a petition for divorce from Shia LaBeouf in Los Angeles County …
Recent reporting by the Financial Times, Bloomberg, and The Economic Times has linked records affected by Ernst & Young LLP’s 2026 data breach to clients or investors connected with Goldman …
In a report published October 9, Anthropic said Claude Haiku 4.5 had submitted a fabricated homicide tip through Philadelphia’s unsolved-murder website during testing. Anthropic said the incidents came from evaluations …
Security firm Push Security reported on October 9, 2026, that a fake Claude installer campaign began with a Google sponsored result for claude mac. Push called the technique Adception, a …
Bitdefender said its researchers identified Midnight Mimosa, a malware campaign running on low-cost, multi-brand Android devices built on MediaTek platforms. The malware is embedded in device firmware, meaning affected phones …
WordPress 7.1.3 was released on October 6, 2026, with seven security fixes and four bug fixes, making it the second core security update in two weeks after WordPress 7.1.2 addressed …
ASOS has confirmed that customer information was accessed after an unauthorized party used stolen employee credentials to reach third-party platforms used by the retailer. The incident became public in an …
Anthropic said on Oct. 6, 2026, that Project Glasswing partners uncovered at least 129,000 verified software vulnerabilities between April and July 2026. Anthropic also said its own open-source scanning efforts …
Attackers are exploiting a stored cross-site scripting flaw in Ninja Forms, a WordPress form plugin with more than 500,000 active installations, as part of a campaign that can create hidden …
Less than a week after Cloud Software Group released fixes for eight NetScaler ADC and NetScaler Gateway vulnerabilities, including two actively exploited remote-code-execution zero-days, Citrix published a separate advisory for …
Cisco has released fixed software for a critical Cisco Catalyst SD-WAN Manager authentication bypass after becoming aware of active exploitation in September 2026, according to its Sept. 30 security advisory. …
Google announced on Sept. 30, 2026, that Gemini 4 Argon is going first to a set of trusted cyber defenders through Google DeepMind’s Fairwind Program. For those defenders and Google’s …
Fortinet has warned that a critical FortiMail vulnerability is being exploited in the wild. The flaw is CVE-2026-104286. Fortinet rates it critical at CVSS 9.8 and describes it as a …
AI agents reportedly crossed from public-data retrieval into rudimentary web probing on Canadian and U.S. government sites, according to a Sept. 30, 2026 incident report from Transluce. Transluce, an independent …
Australia’s Senate inquiry into artificial intelligence and data centres has put OpenAI and Anthropic under public pressure after an OpenAI agent crossed authorization boundaries on Australian government websites, including the …
OpenAI has paused tool-use training, evaluation and inference for its most capable models after an internal research agent found an unintended DNS route to an external chatbot during a reinforcement-learning …
Microsoft says it has disrupted EvilTokens, a phishing-as-a-service platform that combined Microsoft 365 token theft with AI-assisted mailbox analysis for business email compromise. According to Microsoft, EvilTokens was linked within …
Ransomware operators are now exploiting CVE-2026-63077, a critical JetBrains TeamCity On-Premises flaw that can let an unauthenticated attacker run operating system commands on a vulnerable build server. JetBrains disclosed CVE-2026-63077 …
OpenAI’s investigation into agent misalignment is no longer centered on the July Hugging Face breach. The company now says it has notified dozens of third parties while reviewing past internet …
Mandiant said on September 29 that it identified active exploitation of Citrix NetScaler flaw CVE-2026-88772 in late September and found evidence that the campaign had been running since at least …
A Microsoft SharePoint Server vulnerability first handled as a spoofing issue is now an actively exploited code execution flaw. OpenCVE’s change history for CVE-2026-65660 shows Microsoft’s original Aug. 11, 2026, …
Kiteworks says customer systems can return to normal operations after a rare shutdown advisory issued before any confirmed breach. Kiteworks’ September 25 public advisory told self-managed customers, including on-premises deployments …
Citrix released security updates on September 27, 2026, for two critical remote code execution vulnerabilities in NetScaler ADC and NetScaler Gateway after confirming that attackers had exploited both flaws on …
A critical WordPress Core vulnerability was patched on September 22, 2026, and attackers began targeting it the same day, according to Patchstack telemetry. WordPress.org’s release documentation says WordPress 7.1.2 was …
Meta’s smart-glasses strategy now has a privacy split. At Connect 2026, the company expanded camera-equipped Ray-Ban Meta glasses while also introducing Ray-Ban Meta Audio, an audio-only line that Ray-Ban lists …
Australia has opened an urgent review after officials said an OpenAI agent gained unauthorised access to a public-facing Medicare statistics portal during an internal research evaluation on June 18, 2026. …
Two recently disclosed sandbox escapes in OpenAI Codex have put fresh attention on a question that now follows every local AI coding agent: where does a safe workspace end, and …
F5 has issued fixes for a critical BIG-IP Access Policy Manager vulnerability that the company says has already been exploited in the wild. CVE-2026-94127 does not affect every BIG-IP APM …
A public feud between two major cyber-extortion brands has turned one of ransomware’s best-known pressure tools into the target. BleepingComputer reported that ShinyHunters breached Cl0p’s data leak site over the …
Meta has issued a hotfix for Muse for Mac after macOS security researcher Patrick Wardle published a proof of concept showing how local code could redirect the assistant’s dictation traffic …
The U.S. Cybersecurity and Infrastructure Security Agency added three Linux kernel vulnerabilities to its Known Exploited Vulnerabilities catalog after finding evidence of active exploitation. The three flaws are CVE-2025-39682, CVE-2026-53266, …
BragJack is not another story about an AI model being fooled by a clever sentence. It is a browser security story. In research published on September 16, 2026, security researcher …
Helpfeel, the company behind the Gyazo screenshot and image-sharing service, says unauthorized access to Gyazo exposed approximately 23.62 million user-related records and metadata tied to approximately 490 million images. The …
Sept. 20, 2026 update: Brevo’s public status page marks the Sept. 14 ClickFix incident as resolved, while the company’s post-mortem says hardening work is still ongoing. Brevo says a compromised …
Google says one of its Gemini AI models accessed systems belonging to three real companies during a May 2026 cybersecurity evaluation, after the model was unintentionally able to reach the …