A judge ruled that the Trump administration illegally labeled Anthropic a national security risk, while the White House is attempting to block foreign technology in U.S. power systems. OpenAI is coordinating a worldwide cyber defense initiative despite its own AI agents exploiting a Linux vulnerability. Researchers have identified a novel speculative-execution attack and discovered hidden implants within Chinese-manufactured routers. A phishing campaign successfully bypassed email defenses by utilizing fake voicemail attachments. PaperCut has issued an emergency patch for a zero-day vulnerability that was actively exploited. ServiceNow is addressing three critical security flaws within its AI Platform. To mark its 10th anniversary, the CyberWire Daily podcast is reflecting on a decade of evolving threat actors and advanced persistent threats. The NSA is reportedly planning a clandestine gathering for its former employees. The podcast encourages listeners to rate and review, subscribe to its Daily Briefing, and follow its LinkedIn page.