Vanta Launches the Vanta Trusted Network to Turn Compliance Partners Into Recurring GRC Practices
A day after agreeing to buy Pillar Security, Vanta announced a change in how it sells. On November 10 it launched the Vanta Trusted Network,…
AppGate Earns the Only Perfect Score for Zero Trust Architecture Integrity in the GigaOm ZTNA Radar
Zero trust network access has become a crowded checkbox. Most big security platforms now include something they call ZTNA, often as one feature among many.…
Bitdefender Replaces TrafficLight With Web Sage, an AI Browser Advisor That Explains Why a Page Is a Scam
Consumer security used to mean scanning email attachments. The fraud has moved. Fake shops, cloned bank pages and scam ads now sit in search results…
Cyera, Obsidian and Druva Ship New Controls for AI Agents as Security Vendors Race to Govern What Agents Do
Three security vendors announced products on November 10 aimed at the same new problem: AI agents that act inside company systems with real permissions. Each…
Rubrik (RBRK) Adds Surgical Recovery for GitHub and Azure DevOps and Extends Backup to BigQuery and AWS Applications
Rubrik used its Cloud Resilience Summit on November 10 to announce three sets of features in one day. They cover source code in GitHub and…
Attackers Exploit New Flaws Within 24 Hours While 44 Percent of Companies Need Four Days to Patch, Lumen and Azul Find
Two reports released on November 10 measure the same race from opposite sides. Lumen’s threat researchers watched how fast attackers move. Azul surveyed how fast…
Finite State Raises 27 Million Dollars to Help Device Makers Find What Is Really Inside Their Firmware
Every connected product runs on code its maker didn’t write. Cars, infusion pumps, baby monitors, routers and grid equipment all ship with third-party libraries and…
Cisco Puts Claude Mythos Into AI Defense for Red Teaming as Linux Foundation Research Finds AI Writes Code Faster Than Security Can Review It
Two announcements on November 9 describe the same problem from opposite ends. Cisco said at Cisco Live APJC in Melbourne that it will add Anthropic’s…
Vanta Buys Pillar Security to Police AI Agents at Runtime as Orca Finds 99.9 Percent of Fixable AI Flaws Left Open
Vanta has signed a definitive agreement to buy Pillar Security, a two-year-old startup that tests and guards AI agents while they run. Price and terms…
Sentra Raises 150 Million Dollars Led by Capital One Ventures to Control What AI Agents Can Reach in Company Data
Sentra, the New York data security company founded in 2021, closed a $150 million Series C led by Capital One Ventures on November 4. Its…
Abnormal AI Survey Finds 46% of Security Teams Cannot Define Normal Behavior for AI Agents
Security teams have spent a decade learning what normal looks like for a person. They’ve barely started on the software that now acts like one.…
Cyber Threat Pressure Reads 55.2 Elevated as CISA Adds Exploited Flaws at the Fastest Pace Since 2022
The Cyber Threat Pressure gauge reads 55.2, Elevated, for October 2026. CISA’s exploited flaw catalog grew at its fastest three-month pace since 2022, while new critical CVEs stayed unusually low.
Check Point Starts Renting Out Its Firewalls to Midsize Companies as Appliance Sales Slide and Fortinet Surges
Check Point’s new Firewall-as-a-Subscription swaps upfront hardware purchases for a three-year monthly or annual fee, aimed at midsize buyers as its own product revenue falls.
ProjectDiscovery Gives Away Neo Triage, Betting Proven Exploits Beat CVSS Scores for Clearing Vulnerability Backlogs
ProjectDiscovery’s Neo Triage reproduces each reported vulnerability in a sandbox against the live asset and returns a verdict with evidence, and it ships in every Neo tier including free.
Fewer Than 20 Percent of AI Decisions Have a Named Owner as Nuix, Basware and GovSignals Push Audit Trails
Sopra Steria Next finds few AI-driven decisions have an owner or can be replayed, while Nuix, Basware and GovSignals push audit trails and human sign-off as the fix.
AI Code Still Fails 38 Percent of Security Tests as One in Seven US Healthcare Sites Runs Known Exploited Flaws
Veracode says AI-written code now passes 62% of security tests, up from 56%. SecurityScorecard found known exploited flaws on one in seven US healthcare sites. Read together, they point to triage as the real bottleneck.
CISA Adds Exploited VMware ESX Logging Flaw and Fragnesia Linux Root Bug to KEV Catalog
CISA added two exploited flaws to its KEV catalog on October 28: a low-scored VMware ESX logging gap that lets admins act unseen, and the Fragnesia Linux kernel bug that turns a local user into root. Federal agencies get until October 31 for the Linux fix.
ISC2 AI Security Pilot Puts Governance and Model Defense on the Same Exam
ISC2 has published the outline for its forthcoming AI Security Certification pilot, offering a useful picture of the skills employers may soon expect from people…
Your AI Assistant Is Becoming a New Route Into Your Accounts
Connecting an assistant to email, documents and business software changes the account-security picture. The assistant can now retrieve private material and, depending on the integration,…
Tracking Pixels Are a Compliance Liability Most Sites Never Audit
Hospital settlements, FTC cases, VPPA and CIPA suits and European fines all trace back to marketing tags nobody checked. Here is what a real pixel audit looks like.
Datadog Offers Nearly 4 Billion Dollars for Israeli Cloud Security Startup Upwind
Datadog has offered close to $4 billion for Upwind Security, but the Israeli cloud security startup has not accepted and is finalizing a $300 million round at a valuation of about $3.8 billion.
Cloudflare Deploys PHP2Shell Defenses as Critical PHP Flaws Put Websites at Risk
Precautionary website shutdowns spread as organizations race to patch PHP2Shell, while Cloudflare deploys firewall defenses and security agencies urge immediate action.
Cybersecurity Digest, October 19, 2026: AI-Powered Intrusions, Cloud Compromises and Critical Infrastructure Vulnerabilities
The cybersecurity landscape is changing in ways that make traditional distinctions between sophisticated attackers and ordinary cybercriminals increasingly difficult to maintain. Today’s developments illustrate the…
Continuum AI Pushes Autonomous Cybersecurity From Vulnerability Discovery Toward Tested Repairs
Continuum AI has introduced OrcaCyber Zero 1.5, a cybersecurity reasoning model designed to investigate vulnerabilities, validate exploitability and support remediation. The model includes a million-token context…