CyberWire Daily
The AI lock comes off.
· Summarized
Key takeaways
- The U.S. has lifted restrictions on advanced AI models from Anthropic, contingent on security collaborations, reflecting evolving AI regulation dynamics.
- Critical vulnerabilities in enterprise platforms like Adobe and Citrix continue to pose significant risks, demanding prompt patching and vigilance.
- Emerging threats like the Rust Duck botnet and sophisticated ransomware exploits highlight the need for proactive detection and response strategies.
- The space industry is rapidly integrating security considerations, exemplified by Rocket Lab’s potential acquisition of Iridium, aiming for secure satellite communications.
- Browser-only ransomware shows how innovative attack techniques can bypass traditional defenses, necessitating new detection approaches.
- Countries like New Zealand need to enhance cybersecurity readiness to protect critical sectors from increasingly frequent and damaging breaches.
- Expert insights suggest that rebuilding trust online requires cybersecurity measures rooted in relationship verification and security by design, particularly for vulnerable populations.
- AI’s role in security is double-edged, as it accelerates both vulnerabilities and defensive innovations, emphasizing the importance of continuous adaptation.
Summary
U.S. Eases Export Restrictions on Advanced AI Models
The U.S. government has lifted restrictions on anthropics' AI models, Claude Fabel-5 and Mythos-5, allowing Anthropic to resume exports. These models were previously suspended over security concerns relating to vulnerability exploitation potential. The decision followed commitments by Anthropic to enhance security measures and collaborate with government regulators. This move highlights ongoing global debates about AI security and trust.
Cybersecurity Patches and Emerging Threats
Adobe and Citrix have issued critical patches addressing vulnerabilities in their platforms, including remote code execution flaws and memory bugs. Meanwhile, the Rust Duck botnet, written in Rust, is rapidly evolving, capable of launching DDoS attacks and evading detection with advanced anti-analysis techniques. The group The Gentleman has exploited a zero-day driver flaw to disable security tools and deploy ransomware, exemplifying escalating sophistication in threat actors.
Space Industry and Cybersecurity Developments
Rocket Lab plans to acquire Iridium in an $8 billion deal to create a vertically integrated satellite network supporting secure, space-based communications, including in contested environments. Concurrently, researchers demonstrated how browser-only ransomware could encrypt files using HTML5 APIs, raising alarms about new attack vectors. These developments underscore increasing reliance on space infrastructure and innovative cyber threats.
National and International Cybersecurity Concerns
New Zealand faces questions about its cybersecurity readiness following significant breaches in healthcare, highlighting vulnerabilities in critical infrastructure. Meanwhile, arrests related to Iran’s cyber espionage campaigns reveal continued state-sponsored efforts against U.S. and international targets, causing billions in damages and exposing geopolitical cyber tensions.
Expert Interview on Digital Trust and Security by Design
Donald Codling, CISO and strategic advisor, emphasizes that the erosion of trust from AI-generated misinformation and cyber threats necessitates security approaches rooted in trust and relationship verification. He advocates for security by design, especially for vulnerable populations like children and the elderly, and highlights the importance of trusted institutions, such as financial organizations, in establishing online trust frameworks.
Innovative Findings and Final Insights
Research demonstrates how AI can accelerate vulnerability discovery, exemplified by Ian Carroll’s backstage pass incident where AI enabled exploiting a web app firewall and SQL injection to issue fake event tickets. The episode concludes with reminders for listeners to stay informed, rate the podcast, and explore sponsors.
Sponsors
- Black Hat USA · While you're at Black Hat USA, visit the Specterops Kennel Club for talks, workshops, and sessions on AI attack paths, identity tradecraft, and building open graph collectors. Pre-register at Spectrops.io. Also, visit the N2K CyberWire Podcast Studio to catch expert conversations from Black Hat.
- Google Chrome · This episode is brought to you by Google Chrome. It can help you with anything online, from restoring vintage motorcycle restorations to breaking down long articles.
- Vanta · Vanta is the agentic trust platform trusted by over 16,000 companies, helping them stay audit-ready and manage AI and vendor risks with automation and continuous monitoring.
- Indeed Sponsor Jobs · Use Indeed Sponsor Jobs to boost your job posts, reach candidates with the right skills, and reduce hiring time.
- American Express · Apply for the American Express Graphite Business Cash Unlimited card with 2% cashback on all purchases and additional benefits. Earn $1,500 cashback after qualifying spending.
Show notes
The US restores exports of Anthropic’s most advanced AI models. Adobe and Citrix rush out critical patches. RustDuck emerges as a fast-evolving DDoS threat. The Gentlemen raise the stakes with a new EDR-killing exploit. Rocket lab bets big on Iridium. Researchers unveil browser-only ransomware. New Zealand faces questions about its cyber readiness. Iran’s long-running cyber espionage campaign is back in the spotlight. Our guest is Donald Codling, CISO and senior advisor to REGO on cybersecurity and data privacy matters, to discuss the importance of tying security by design to psychological safety and digital trust. VIP backstage access, courtesy of Claude.
Remember to leave us a 5-star rating and review in your favorite podcast app.
Miss an episode? Sign-up for our daily intelligence roundup, Daily Briefing, and you’ll never miss a beat. And be sure to follow CyberWire Daily on LinkedIn.
CyberWire Guest
Today we are joined by Donald Codling, CISO and senior advisor to REGO on cybersecurity and data privacy matters, to discuss the importance of tying security by design to psychological safety and digital trust.
Selected Reading
Fable and Mythos: Anthropic says US lifts export ban on its advanced AI tools (BBC)
Adobe patches seven max severity ColdFusion, Campaign flaws (Bleeping Computer)
RustDuck: The Botnet That's Still Small but Engineering Like It Plans to Grow (SecurityAffairs)
Citrix Patches NetScaler Vulnerabilities, Including New ‘HTTP/2 Bomb’ Attack (SecurityWeek)
Not very gentlemanly: Analyzing a zero-day exploit used by The Gentlemen ransomware to disable targets’ EDRs (Expel)
Rocket Lab to Acquire Iridium in Historic Deal, Creating A Fully Vertically Integrated Space Powerhouse Primed for Growth (Globe Newswire)
Ransomware that runs inside your browser tab, where antivirus cannot see it (Suriq)
Three major cybehttps://suriq.io/blog/browser-only-ransomware-file-system-accessrattacks have raised alarms about New Zealand's security (RNZ)
Arrest of Iranian Hacker Spotlights Iran’s Movement into Economic Espionage and IP Theft (Zero Day)
Claude Helped a Hacker Find a Way to Issue Tickets to Almost Every US Music Festival (WIRED)
Share your feedback.
What do you think about CyberWire Daily? Please take a few minutes to share your thoughts with us by completing our brief listener survey. Thank you for helping us continue to improve our show.
Want to hear your company in the show?
N2K CyberWire helps you reach the industry’s most influential leaders and operators, while building visibility, authority, and connectivity across the cybersecurity community. Learn more at sponsor.thecyberwire.com.
The CyberWire is a production of N2K Networks, your source for strategic workforce intelligence. © N2K Networks, Inc.
Learn more about your ad choices. Visit megaphone.fm/adchoices
Follow this show on AudioBrevity and new episodes get summarized for you. Create a free account