On July 1st, 2026, the Trump administration reverses Anthropic's export ban, Anthropic doubles down with Claude Sonnet 5 and a scientific research platform, Etched hits a $5B valuation, Google floods the budget tier, and researchers show that lying to an AI browser is enough to break its guardrails.

Sources:

Trump Lifts Export Controls on Anthropic's Mythos and Fable AI Models After Two-Week Ban (Washington Post)
techcrunch.com
wired.com
theverge.com
Anthropic Launches Claude Sonnet 5 and Claude Science Research Platform (TechCrunch)
techcrunch.com
technologyreview.com
Nvidia Rival Etched Hits $5B Valuation with $1B in Contracted AI Chip Sales (TechCrunch)
Google Releases Nano Banana 2 Lite and Gemini Omni Flash for Faster, Cheaper AI Generation (Google DeepMind Blog)
techcrunch.com
China-Linked Actors Escalate AI Espionage Targeting U.S. Startups and Research (CNBC)
AI Browser Security Flaw: Telling an LLM '2+2=5' Bypasses Safety Guardrails (Ars Technica)
Rocky Week for AI Stocks: Markets Sour but No Crash Yet (The Guardian)
cnbc.com
Google NotebookLM Adds TikTok-Style 60-Second AI Video Clips (The Verge)
Netflix Uses AI-Generated Gene Wilder Voice in Willy Wonka Reality Show (The Verge)
Proton Launches Lumo 2.0 with Image Generation, Memory, and Private Web Search (9to5Mac)
techcrunch.com