Global Scans · Security · Weekly Summary

• [New] AWS and NVIDIA will build secure AI infrastructure for US federal and national security workloads. Fintech Singapore
• [New] Given Mythos's impact in shifting the Overton window on AI policy, the Trump administration may become more willing to implement serious oversight measures as they receive further concrete evidence of significant national security risks. Institute for Law & AI
• [New] Various major vendors are enabling GNSS chips with Galileo, and emerging nations are anticipated to focus businesses on enabling chips with Galileo for security purposes. Straits Research
• [New] Israel has understandably focused on immediate security threats. Technocracy News
• [New] The UAE is responding to a genuine security threat after attacks on shipping, infrastructure, and vessels connected to the Abu Dhabi National Oil Company. Armstrong Economics
• [New] The US-Israeli war with Iran has proved more consequential for global energy security, trade and international stability than initially anticipated, while the US military intervention in Venezuela challenged established principles of national sovereignty and international law. Modern Ghana
• [New] PRC cyberattacks pose an ongoing risk to the security of the United States. American Enterprise Institute - AEI
• [New] OpenAI, Anthropic, Google and Microsoft are among the AI builders warning about increasingly capable AI attacks, while security heavyweights including Cloudflare, CrowdStrike, Fortinet, and Palo Alto Networks have also signed on. theregister
• [New] Cybersecurity protection for critical infrastructure similarly depends heavily on European regulatory frameworks such as the Network and Information Security 2 Directive, while NATO identifies operational vulnerabilities and resilience requirements. Hudson Institute
• [New] In the short term, European security will continue to rely on American capabilities and industrial support, while strategic autonomy remains a long-term project whose success depends on Europe's ability to reindustrialize before the strategic window closes. Hudson Institute
• [New] Additional movement restrictions may be imposed on U.S. government employees working in Israel, with little to no notice, due to increased security issues or threats. U.S. Embassy Jerusalem
• [New] As Google's documentation states, manual sign-ins during device setup does not just create onboarding friction and decrease user retention, it actively exposes apps to critical security vulnerabilities like phishing and credential theft during setup. theregister
• [New] Global security awareness training spend will reach $10 billion annually by 2027, up from approximately $5.6 billion in 2023, on roughly 15% year-over-year growth. Swif.ai
• [New] Europe has increased defence spending and accelerated efforts to strengthen its military-industrial capacity because it can no longer assume that the United States will always carry the principal burden of European security. Modern Ghana
• [New] If European allies continue to strengthen their defence capabilities and make tangible progress toward the transatlantic burden shift, they will reinforce the case that NATO remains indispensable to American security and prosperity. The Ronald Reagan Presidential Foundation & Institute
• [New] Security Command Center provides centralized visibility into vulnerabilities and misconfigurations, while Web Security Scanner can help identify common web application vulnerabilities in supported applications. Maze
• Security operations center (SOC) teams and threat intelligence analysts get to see how China -, Russia -, DPRK -, and Iran-aligned actors operate day to day, from AI-assisted exploit development to C & C hidden inside trusted cloud services and blockchains. TrendAI
• In the 2026 Thales Data Threat Report, 70% of organizations rank AI as top data security risk and 48% have experienced reputational damage as a result of AI-generated misinformation. Thales Cloud Security Products
• The establishment of the DE-AISI is an important opportunity for Germany to contribute to the safety and security of frontier AI models as their risks for national security and critical infrastructure become increasingly central. Institute for Law & AI

Last updated: 15 September 2026