Cybersecurity Today

AI agents are changing both sides of cybersecurity. Attackers can move faster, while defenders face another challenge: securing the agents inside their own organizations. On this Cybersecurity Today Weekend Interview, I'm joined by Adam Ely, General Manager of AI Agent Security at Check Point Software. We talk about what AI is changing for security teams—and what still comes down to getting the basics right. Adam brings the perspective of a former CISO at Fidelity Investments and deputy CISO at Walmart. We discuss the pressure on security leaders, the promise of AI for defenders, and how organizations can adopt agents while controlling what they can access and do. In this conversation: • Why AI changes the speed and scale of cyberattacks. • CISO burnout, accountability, and the value of sharing experiences. • How AI agents can influence one another's behaviour. • Why security teams need to rehearse and test AI-assisted incident response. • The identity and access challenges of agents that can create other agents. • What it means when companies put agents on the organizational chart. • Why segmentation, access controls, monitoring, and other security basics matter more than ever. • Using AI as a sounding board that challenges your thinking. Where could AI help your security team—and where would you draw the line on its access? Share your thoughts in the comments. Hosted by Jim Love. Cybersecurity Today — Weekend Interview October 31, 2026 Subscribe to Cybersecurity Today for cybersecurity news and interviews. More episodes and contact: https://technewsday.com https://technewsday.ca To support the show, visit the Donate tab. Guest biography, Check Point Software: https://www.checkpoint.com/about-us/leadership/ Check Point Software research discussed in the interview: https://blog.checkpoint.com/security/peer-pressure-ai-agents Adam's article on the changing CISO role: https://blog.checkpoint.com/ai-security/redefining-the-ciso-contract-from-securing-the-business-to-securely-doing-business/

Black October in Cybersecurity: PHP RCE Panic, Brotli Surprise, and Why Basics Still Matter In this special episode of Cyber Security Today, Jim and David discuss why the industry is calling the month "Black October," beginning with chaotic disclosure and leaked details of multiple PHP remote code execution vulnerabilities that triggered widespread anxiety and even service shutdowns by governments and other organizations. They cover additional tensions from weak certificates issued by an intermediate CA in India, a breach disclosure by LiteLLM, and a major RCE initially thought to affect React/Node.js but later tied to Google's Brotli, which is widely deployed. They note that while attacks began before patches and at least one European government was compromised, exploitation proved difficult and defenses often held, with limited confirmed victims so far. They argue AI is accelerating vulnerability discovery and democratizing capabilities, but emphasize "don't panic," focus on cyber hygiene basics, prioritize crown jewels, train staff, and address growing human fatigue as incident response capacity strains 00:00 Black October Begins 00:44 PHP Zero Day Chaos 03:51 More Hits Keep Coming 04:53 AI Hype And Reality 07:01 Exploitation In The Wild 09:15 Japan Breach Wave 13:23 Dont Panic Do Basics 15:38 Turn Crisis Into Budget 18:07 Training Beats Tools 21:21 Burnout On The Frontline 23:45 Know Your Crown Jewels 24:16 Final Thoughts And Outreach

Brotli Emergency Patches, PHP-to-Shell Exploited, and Australia Disrupts 180K Amadey Infections | Cybersecurity Today On Cybersecurity Today (Oct. 28, 2026), Microsoft issues out-of-band .NET updates for a critical flaw in Google's Brotli library (CVE-2026-111497) that can enable remote privilege escalation and potentially remote code execution, with proof-of-concept exploits public and widespread exposure across ecosystems. The episode also covers CISA confirming active exploitation of two "PHP to Shell" vulnerabilities (CVE-2026-109618/109619), with federal agencies facing an urgent patch deadline and broad impact across millions of PHP services and many vendors. In legal news, U.S. prosecutors allege Canadian ransomware negotiator Eduard Dubrovsky secretly worked for threat actors. South Korean police say a former penetration tester used the Artex AI framework to breach 90+ organizations and steal 8.2 million records. Finally, Australia's Signals Directorate disrupts an Amadey botnet, cutting off 180,000 infected computers and destroying stolen data offered for sale. 00:00 Top Stories Rundown 00:32 Brotli Bug Emergency Patches 02:36 AI Noise in Vulnerability Hype 03:41 PHP to Shell Exploited Now 06:10 Ransomware Negotiator Accused 07:41 Artex AI Hacking Spree 10:04 Australia Disrupts Amadey Botnet 11:32 Wrap Up and Sign Off

PHP2Shell Hits European Government, New PHP RCE (No Patch Yet), Brotli "Node to Shell," and Capcom Breach Attackers exploited the PHP2Shell remote code execution flaws to compromise a European government entity within two hours, stealing password hashes and MFA secrets, pivoting via NTLM relay toward Active Directory, and showing command output patterns Unit 42 says suggest an agentic AI attacker. Widespread scanning continues, with Cloudflare logging millions of attempts, Cyfirma reporting low-and-slow probing from nodes linked with moderate confidence to the Lazarus ecosystem, and defenders urged to watch for telltale crash sequences. PHP also disclosed a third RCE (CVE-2026-115560) with no patch yet but high exploitation requirements and a mitigation to disable session upload progress. Separately, a Brotli decoder memory corruption bug (CVE-2026-111497) affects Node.js and other software, with fixes delayed into October releases. Finally, Capcom says criminals breached Capcom ID, exposing data tied to 11.7 million accounts and resetting impacted passwords. 00:00 Headlines Overview 00:35 Government Breach Details 01:13 Global Scanning Surge 03:11 Third PHP RCE 05:00 Wildfire Or Smoke 06:08 Patch Race Warning 06:53 Brotli Node To Shell 08:47 Capcom Account Breach 10:21 Wrap Up And Outro

PHP Exploited Before Patch, Unpatched Linux SMB Bugs, GitLab Max-Severity Fix, and AI Agents in ConnectWise & Japan Breaches Cybersecurity Today (Oct. 23, 2026) covers active exploitation of two critical PHP RCE flaws beginning at least four days before patches shipped, with multiple firms seeing failed pre-disclosure attempts and the UK NCSC urging immediate patching or taking servers offline. It also details two publicly disclosed, unpatched Linux kernel ksmbd SMB RCE bugs (one unauthenticated on port 445), exposed after ZDI credited Trend Micro's AI scanner and copied public mailing lists without patches. GitLab issued fixes for a maximum-severity unauthenticated RCE (CVE-2026-11-0264) in its regex compiler, following recent high-severity GitLab issues. Microsoft attributed the ConnectWise cloud backup compromise to China-aligned Storm-0940 using the Artex agentic AI framework, and Artex-related activity was also found in attacks on Japanese companies amid a broader breach wave, including a malware incident affecting up to 26,739 Haname Desktopcloud customer servers. 00:00 Headlines And Intro 00:33 PHP To Shell Exploitation 02:07 Patch Now Impact Scope 02:47 Linux KSMBD Bugs Exposed 04:10 AI Disclosure Process Drama 05:25 GitLab Max Severity Patch 07:30 ConnectWise Breach AI Agents 09:24 Artex Spreads In Japan 11:18 GMO Haname VPS Malware 12:24 Closing And Sign Off