Latest / Daily Cyber Briefing / Daily Cyber & AI Briefing — 2026-09-22

Transcript

Daily Cyber & AI Briefing — 2026-09-22 · Daily Cyber Briefing

• 0:07
  SPEAKER_00: Grab your coffee or Red Bull or whatever your morning vice is, and this is your daily cyber and AI briefing, and I am your host, Michael Hoosh.
• 0:16
  SPEAKER_00: Today's cyber and AI risk landscape is evolving at an unprecedented pace, with both technological advancements and threat activity accelerating simultaneously.
• 0:26
  SPEAKER_00: The convergence of these trends is presenting new challenges for organizations from the boardroom to the Security Operations Center.
• 0:36
  SPEAKER_00: Let's break down the most significant developments shaping today's risk environment and what they mean for security leaders, risk executives, and organizations at large.
• 0:47
  SPEAKER_00: Global attention on artificial intelligence risk is reaching a new level.
• 0:52
  SPEAKER_00: The United Nations Security Council is convening high profile briefings with major AI stakeholders, including DeepSeq and Sam Altman, the CEO of OpenAI.
• 1:03
  SPEAKER_00: This is more than a symbolic gesture.
• 1:05
  SPEAKER_00: It's a clear sign that AI governance has become a matter of international security and diplomacy.
• 1:11
  SPEAKER_00: As AI capabilities continue to advance and geopolitical competition intensifies, global leaders are recognizing the need for coordinated frameworks to address AI misuse, safety, and the cross-border impacts of these technologies.
• 1:28
  SPEAKER_00: This growing focus at the UN level is likely to drive new regulatory expectations for organizations deploying advanced AI.
• 1:37
  SPEAKER_00: Security leaders should anticipate that internal AI governance will need to align with emerging international standards, not just national or industry specific requirements.
• 1:49
  SPEAKER_00: This is about more than compliance.
• 1:52
  SPEAKER_00: It's about demonstrating responsible stewardship of AI, ensuring transparency, and preparing for a world where AI related incidents can have global consequences.
• 2:02
  SPEAKER_00: At the same time, research is exposing a significant gap between the rapid adoption of AI in the enterprise and the maturity of governance practices.
• 2:13
  SPEAKER_00: A new EMA research webinar is set to examine this issue in depth, highlighting that while AI is being integrated into business processes at breakneck speed, controls around risk, compliance, and security are struggling to keep up.
• 2:28
  SPEAKER_00: The implications are serious.
• 2:30
  SPEAKER_00: Operational, reputational, and regulatory risks are mounting for our organizations that lack dedicated AI governance frameworks and cross-functional oversight.
• 2:41
  SPEAKER_00: This governance gap isn't just theoretical.
• 2:44
  SPEAKER_00: According to recent Asaka research, only 8% of organizations conduct regular AI-specific incident response exercises.
• 2:53
  SPEAKER_00: That means more than nine out of ten enterprises are not actively preparing for AI-driven attacks or failures.
• 3:00
  SPEAKER_00: As AI systems increasingly power critical decision making and automation, this lack of preparedness is a glaring vulnerability.
• 3:09
  SPEAKER_00: Security and risk leaders need to prioritize scenario planning and tabletop exercises that address the unique risks posed by generative and agentic AI systems.
• 3:20
  SPEAKER_00: Let's talk about what those risks look like in practice.
• 3:25
  SPEAKER_00: One area of concern is the potential for unbounded resource consumption in large language models, a risk highlighted in the OYS PLLM top 10.
• 3:35
  SPEAKER_00: A recent simulation demonstrated how attackers can exploit these models to trigger excessive compute or data usage.
• 3:42
  SPEAKER_00: The result?
• 3:43
  SPEAKER_00: Denial of service attacks or runaway costs that can disrupt operations and hit the bottom line.
• 3:51
  SPEAKER_00: Security teams must implement usage controls and monitoring for AI workloads, ensuring that resource consumption is predictable and contained.
• 3:59
  SPEAKER_00: The evolution of AI is also driving a shift in how organizations need to approach governance.
• 4:05
  SPEAKER_00: Akamai's latest report underscores that agentic AI systems capable of autonomous action require a move away from purely technical controls toward behavioral governance.
• 4:18
  SPEAKER_00: This means organizations must monitor AI decision making, enforce guardrails, and ensure accountability at every stage of the AI life cycle.
• 4:27
  SPEAKER_00: Traditional security controls aren't enough.
• 4:30
  SPEAKER_00: It's about understanding and shaping the behavior of AI agents in real-world environments.
• 4:36
  SPEAKER_00: Compounding this, agentic AI systems are not exempt from existing compliance requirements.
• 4:43
  SPEAKER_00: Cybersecurity Insiders points out that frameworks like the Cybersecurity Maturity Model Certification and state privacy laws still apply.
• 4:53
  SPEAKER_00: Organizations can't afford to treat AI as a special case outside the scope of established controls.
• 5:00
  SPEAKER_00: Instead, they need to integrate AI-specific measures into their broader compliance and privacy programs, especially as regulatory scrutiny intensifies.
• 5:10
  SPEAKER_00: While AI risk and governance are quickly rising to the top of the agenda, the cyber threat landscape remains as active as ever.
• 5:18
  SPEAKER_00: Several critical vulnerabilities and breaches have emerged that demand immediate attention.
• 5:23
  SPEAKER_00: First, a coordinated attack has compromised 65 GitHub repositories and introduced a hidden backdoor into an NPM package.
• 5:32
  SPEAKER_00: This is a textbook supply chain attack demonstrating the ongoing risk of dependency poisoning.
• 5:38
  SPEAKER_00: For organizations that rely on open source components, and that's virtually everyone, rigorous code provenance checks and supply chain security practices are essential.
• 5:48
  SPEAKER_00: It's not just about vetting your own code, but also understanding and managing the risks introduced by third-party dependencies.
• 5:56
  SPEAKER_00: Meanwhile, Chinese advanced persistent threat groups are leveraging cloned legitimate websites to deliver Chrome and Windows zero-day exploits.
• 6:06
  SPEAKER_00: This tactic raises the risk of drive-by compromise, where unsuspecting users are infected simply by visiting a trusted-looking site.
• 6:14
  SPEAKER_00: The practical takeaway here is the importance of user awareness, robust patch management, and integrating threat intelligence into security operations.
• 6:25
  SPEAKER_00: Organizations need to ensure that their endpoints are patched quickly and that users are trained to recognize suspicious activity, even when it appears to come from familiar sources.
• 6:36
  SPEAKER_00: On the vulnerability management front, the Cybersecurity and Infrastructure Security Agency, KISA, has added six new exploited flaws to its known exploited vulnerabilities catalog.
• 6:50
  SPEAKER_00: Federal agencies have been given just three days to patch, reflecting the accelerating pace at which vulnerabilities are being weaponized in the wild.
• 6:58
  SPEAKER_00: This is not just a federal issue.
• 7:01
  SPEAKER_00: The expectation for rapid vulnerability management is becoming the norm across all sectors.
• 7:07
  SPEAKER_00: Organizations need to maintain up-to-date asset inventories, automate patch deployment where possible, and ensure that both IT and operational technology environments are covered.
• 7:20
  SPEAKER_00: The risks associated with identity and authentication providers have also been underscored by a massive breach at Nexus, which exposed 153 million identities.
• 7:31
  SPEAKER_00: The sheer scale of this incident raises significant counterintelligence and fraud risks, not just for Nexus, but for every organization that relies on third-party authentication.
• 7:43
  SPEAKER_00: This highlights the need for robust third-party risk management, layered authentication controls, and continuous monitoring for signs of credential abuse or identity compromise.
• 7:54
  SPEAKER_00: Another area of immediate concern is the exploitation of zero-day vulnerabilities and network infrastructure.
• 8:01
  SPEAKER_00: A zero-day exploit chain targeting SonicWall SMA 1000 appliances has been rated with a maximum CVSS score of 10.0, indicating the highest level of severity.
• 8:13
  SPEAKER_00: Organizations using these devices should move quickly to patch, but also consider additional measures like network segmentation and enhanced monitoring to detect and contain any potential compromise.
• 8:24
  SPEAKER_00: Similarly, CESA has issued an alert for an actively exploited vulnerability in Zixel GS1900 switches.
• 8:32
  SPEAKER_00: Network infrastructure devices remain high value targets for attackers seeking persistent access or lateral movement within organizations.
• 8:41
  SPEAKER_00: Timely patching and ongoing monitoring are critical to defend against these threats.
• 8:46
  SPEAKER_00: Stepping back, what are the strategic implications of these developments?
• 8:51
  SPEAKER_00: First, international AI governance is accelerating.
• 8:55
  SPEAKER_00: Organizations deploying advanced AI should expect regulatory expectations to increase and should begin aligning their internal policies with emerging global standards.
• 9:06
  SPEAKER_00: This isn't just about checking boxes.
• 9:09
  SPEAKER_00: It's about building trust with stakeholders, customers, and regulators in an environment where AI risks are under the microscope.
• 9:16
  SPEAKER_00: Second, the gap between AI adoption and governance is a critical risk.
• 9:22
  SPEAKER_00: Most organizations are not adequately prepared to respond to AI-specific incidents or to govern the behavior of autonomous systems.
• 9:30
  SPEAKER_00: This is now a board level issue, requiring investment in dedicated governance frameworks, cross-functional oversight, and ongoing training and exercises.
• 9:41
  SPEAKER_00: Third, supply chain and identity breaches are escalating.
• 9:45
  SPEAKER_00: Organizations must double down on third-party risk management, robust authentication controls, and continuous monitoring for compromise.
• 9:54
  SPEAKER_00: The risks introduced by partners and providers are no longer peripheral.
• 9:58
  SPEAKER_00: They are central to organizational security.
• 10:00
  SPEAKER_00: Finally, the pace of vulnerability exploitation is only increasing.
• 10:05
  SPEAKER_00: Rapid patch cycles, improved asset visibility and automation are essential to keep up with the threat landscape.
• 10:12
  SPEAKER_00: This applies not just to traditional IT assets, but to operational technology and network infrastructure as well.
• 10:20
  SPEAKER_00: So, what matters most today for security and risk leaders?
• 10:24
  SPEAKER_00: First, prepare for new AI governance requirements by proactively aligning your internal policies and controls with international standards.
• 10:32
  SPEAKER_00: Don't wait for regulations to be finalized.
• 10:34
  SPEAKER_00: Start building the foundations now.
• 10:36
  SPEAKER_00: Second, accelerate your AI-specific incident response planning.
• 10:41
  SPEAKER_00: This means developing and running tabletop exercises that address the unique risks posed by generative and agentic AI.
• 10:48
  SPEAKER_00: Make sure your teams know how to respond to AI-driven incidents from data leakage to autonomous system failures.
• 10:56
  SPEAKER_00: Third, prioritize patching of newly disclosed vulnerabilities.
• 11:01
  SPEAKER_00: The window between disclosure and exploitation is shrinking and attackers are moving faster than ever.
• 11:06
  SPEAKER_00: Automate where you can and make sure you have visibility in all assets, including those in the cloud and in operational environments.
• 11:14
  SPEAKER_00: Fourth, monitor for supply chain compromise.
• 11:18
  SPEAKER_00: This is especially important for organizations that rely on open source software or third-party authentication providers.
• 11:25
  SPEAKER_00: Implement rigorous code provenance checks, monitor for unusual activity in your software supply chain, and ensure that your identity and access management controls are robust and adaptive.
• 11:37
  SPEAKER_00: And finally, recognize that the operational tempo for cyber defense remains high.
• 11:42
  SPEAKER_00: The convergence of AI governance challenges and escalating cyber threats requires a dual focus.
• 11:49
  SPEAKER_00: Maturing your AI risk management capabilities while maintaining relentless vigilance on patching, supply chain security, and incident response.
• 11:58
  SPEAKER_00: The risks are real and the stakes are high.
• 12:01
  SPEAKER_00: But with the right strategies, governance, and operational discipline, organizations can navigate this evolving landscape and build resilience against both current and emerging threats.
• 12:12
  SPEAKER_00: That's today's cyber and AI risk briefing.
• 12:15
  SPEAKER_00: Stay vigilant, stay informed, and keep building for resilience.
• 12:19
  SPEAKER_00: That's a wrap, peeps.
• 12:21
  SPEAKER_00: Stay secure, stay sharp, and don't forget to hug your CISO.