Major developments in AI regulation, space technology, cybersecurity threats, and scientific policy dominate today's science news, with government orders affecting AI models, SpaceX's public valuation, and critical security vulnerabilities affecting organizations worldwide.

Anthropic has shut down its Fable and Mythos AI models following a Trump administration directive, with the government citing discovered methods to jailbreak Fable 5, marking a significant government intervention in AI development.

SpaceX is now a publicly traded company valued significantly for its AI potential rather than just space launch capabilities, signaling investor focus on the company's broader technological ambitions beyond traditional rocket operations.

A critical PeopleSoft zero-day vulnerability is actively exploiting hundreds of organizations and stealing gigabytes of sensitive data, representing a major cybersecurity threat to enterprise systems worldwide.

Jeff Bezos' new startup Prometheus has been announced, though details remain limited, adding to the billionaire's portfolio of ventures beyond Amazon and Blue Origin.

Underground fungal networks span distances theoretically long enough to reach beyond the Solar System, highlighting the vast scale of mycological ecosystems beneath Earth's surface.

A federal court has ruled that Google is legally liable for false statements generated by its AI Overviews, establishing precedent that companies designing and operating AI systems must assume responsibility for inaccurate AI-generated content.

Ukraine has demonstrated autonomous drone warfare capability in a one-time test using fully self-directed drones to kill Russian soldiers, marking a significant milestone in military AI deployment.

The controversial FISA spying law expires tonight, though surveillance operations are expected to continue, reflecting ongoing tensions between national security and privacy concerns.

The National Academies of Science face questions about political influence on scientific independence and research priorities, suggesting erosion of institutional autonomy.

AI agents are being developed by Spanish researchers to protect EV charging infrastructure from energy theft and damage, demonstrating AI applications in critical energy security.