Full Summary
This Tuesday morning, the U.S. government has ordered Anthropic to suspend global access to its new AI models, Fable 5 and Mythos 5, citing national security concerns. Both The Standard (HK) and CUToday confirm this directive came from U.S. Commerce Secretary Howard Lutnick, who fears the models could be misused by military intelligence in countries like China and Russia. NBC News and 24/7 Wall St. report that Anthropic took its systems offline globally by June 12th, just days after Fable 5's launch, because it could not restrict access to only U.S. users. Fortune adds that Amazon researchers found a way to "jailbreak" Fable's cybersecurity safeguards, prompting Amazon CEO Andy Jassy to contact the White House. While Anthropic claims the bypass only exposed minor flaws, the New York Post notes the Trump administration is open to negotiating a resolution. Meanwhile, a U.S. federal judge has dismissed Elon Musk's xAI trade secret lawsuit against OpenAI for the second time. Crypto News Australia, Silicon UK, and en.ain.ua all report that Judge Rita Lin found xAI failed to provide sufficient evidence that OpenAI improperly induced a former xAI engineer to reveal trade secrets about the Grok chatbot. The judge noted that discussing past work during recruitment is standard and the allegations did not prove OpenAI encouraged disclosure. This dismissal is a significant victory for OpenAI, following another legal loss for Musk against the company last month. In other AI news, OpenAI itself is facing significant financial pressure. Futurism and Gizmodo reveal the company lost around $39 billion in 2025, a sharp increase from $5 billion in 2024. This massive cash burn is leading OpenAI to consider shifting from monthly subscriptions to a token-based billing system, where users pay more directly for computing power. This means that the advanced AI tools you use could become more expensive and subject to stricter government controls, directly impacting your access and the cost of AI-powered services.