LinkedIn tests peer verification for work history
LinkedIn is testing a system where connections can vouch for each other's work and education history, adding a verification badge after enough confirmations.
Security & Privacy
Breaches, offensive and defensive AI, and surveillance.
Anthropic's AI sprint makes Claude 3x faster without breaking anything
A two-week AI-assisted optimization drive cut Claude's latency dramatically while avoiding customer-facing incidents, showing how careful automation can improve performance without compromising stability.
ClickFix Is Now the Leading Enterprise Intrusion Vector, Report Finds
A new CTM360 report traces ClickFix's rise from a 2023 novelty to the most common initial-access technique, with blockchain-based infrastructure that defeats domain blocking.
Phishing URL Packs Three Evasion Tricks Into One Link
A SANS ISC analysis shows how a single phishing link can combine multiple obfuscation techniques to slip past basic security controls.
doxx.net Raises $38M for AI Agent Safety Platform
The funding backs a platform designed to stop AI agents from causing harm while operating on a user's behalf.
Cybersecurity in 2026: From Point Tools to Continuous Risk Control
A new industry report shows security shifting from isolated products toward integrated, continuous visibility across identity, telemetry, endpoints, and human risk.
Microsoft's X Account Hijacked to Push Crypto Token
The compromise of Microsoft's official X account, with over 13 million followers, underscores how social media account takeovers remain a favored vector for crypto scams.
Browser Attacks Slip Past EDR Telemetry, NordLayer Warns
Endpoint detection tools miss browser-based attacks that steal sessions, abuse extensions, or manipulate users, leaving a gap that browser-level controls may help close.
Vicksburg, Mississippi Shuts Down Systems After Ransomware Attack
The city confirmed a ransomware incident that disrupted utility payments but left emergency services operational, with investigators still assessing potential data exposure.
GitLab Patches Critical 9.9 AI Gateway Flaw for Self-Hosted Users
A critical flaw in GitLab's AI Gateway could let authenticated users execute commands on self-hosted gateways, and GitLab is urging immediate updates.
Weekly infosec roundup: AI security, access control, app hardening
New releases from BlackFog, Genea, Vega, and Thales target distinct security challenges, from AI-driven attacks to AI-assisted workflows.
Proton Drive adds encrypted docs and sheets to cloud storage
Proton's end-to-end encrypted storage now includes collaborative document and spreadsheet tools, bringing privacy to productivity.
Employment scams triple at financial institutions in 21 countries
A new fraud report finds employment scams are the fastest-growing scam type, with 90% of scam sessions now starting on mobile devices.
BH Haven pairs AI analysis with human sign-off for SME security
A new Irish service targets small firms' compliance gaps by letting consultants make final calls on AI-assisted assessments.
Attackers Abuse Legitimate ScreenConnect Client
SANS ISC warns that attackers are increasingly abusing the legitimate ScreenConnect remote access tool instead of relying on complex malware.
Old Critical Flaws Dominate Backlogs, Public Sector Slowest to Patch
Detectify's customer snapshot finds most open critical and high-severity vulnerabilities are more than 90 days old, with public bodies resolving the smallest share.
Android 17 Advanced Protection adds tamper-resistant forensic logging
Google's expanded Android 17 protections give targeted users a way to preserve evidence of spyware attacks even if attackers erase on-device traces.
Dell Patches Critical CSM Flaws Enabling Kubernetes Cluster Takeover
Dell's Container Storage Modules have multiple critical vulnerabilities, including two maximum-severity flaws, that can give attackers admin control over storage and Kubernetes clusters.