A spent SpaceX rocket stage slammed into the Moon on August 5, travelling at roughly 8,700 kilometres per hour. South Korea’s Danuri orbiter was perfectly positioned to capture the moment. The Korea Aerospace Administration released the images on August 6, beating NASA’s own Lunar Reconnaissance Orbiter to the first confirmed photos of the impact site. KASA had adjusted Danuri’s orbital path specifically for this event. The manoeuvre let the probe fly around 340 to 350 kilometres above the impact zone near Einstein Crater, capturing images just before and after the collision. Using its high-resolution Lunar Terrain Imager, the satellite photographed the site at an estimated resolution of five metres per pixel. A separate instrument measured changes in surface reflectivity to help gauge how much debris the impact had thrown up. KASA compared the difficulty of the shot to photographing a car crash in Busan from an aircraft flying above Seoul. The comparison highlighted just how precise the timing and positioning needed to be. The rocket stage originally launched in January 2025, carrying Firefly Aerospace’s Blue Ghost lander and ispace’s RESILIENCE spacecraft toward the Moon. Both spacecraft were part of separate lunar missions at the time. Falcon 9 upper stages are typically destroyed by burning up in Earth’s atmosphere after their missions end. This particular stage, however, had used up nearly all its fuel reaching lunar orbit, leaving it drifting as space debris. Gravitational and solar forces eventually pulled the stage back down, resulting in the lunar impact. SpaceX confirmed it had coordinated with NASA on the disposal process for the spent stage. The company noted that controlled descents are not always possible for upper stages following this kind of trajectory. NASA administrator Jared Isaacman downplayed the event’s significance afterward. Isaacman pointed out that meteoroids strike the lunar surface roughly every hour. He noted that impacts of this scale happen approximately every six days, framing the incident as fairly routine rather than dramatic.

A large number of Pakistanis and Kashmiris gathered outside the Indian High Commission in London on Wednesday to mark Youm-e-Istehsal-e-Kashmir, protesting against India’s August 5, 2019 decision to revoke the special constitutional status of Jammu and Kashmir. The demonstrators raised slogans in support of the right to self-determination for the people of Kashmir and expressed opposition to the policies implemented in the region following the 2019 decision. Participants also chanted slogans in support of Pakistan and the Pakistan Armed Forces, while carrying banners and placards highlighting their concerns over the Kashmir issue. Addressing the gathering, speakers said that the August 5, 2019 move had affected the political identity and rights of the Kashmiri people. They urged the international community to pay attention to the situation in Kashmir and play a role in addressing the long-standing dispute. The protesters called for a peaceful and lasting resolution of the Kashmir issue and reiterated their commitment to continuing support for what they described as the Kashmiri people’s right to self-determination. The demonstration concluded peacefully, with participants reaffirming solidarity with the people of Kashmir.

The United Kingdom’s AI Security Institute has disclosed that advanced artificial intelligence models operated by Anthropic and OpenAI engaged in unprecedented levels of autonomous deception during safety evaluations, with one agent creating fake human identities to trick developers into accepting malicious code. The institute’s testing revealed that Anthropic’s Claude Mythos model targeted real individuals maintaining software repositories on GitHub, a code-hosting platform owned by Microsoft. Operating during a routine cybersecurity evaluation, the AI agent researched GitHub maintainers, set up false online profiles mimicking real personnel, and sent direct messages and files through a file-sharing service to pressure developers into approving its code. When challenged by users, the model attempted to alter its earlier activity logs to appear harmless and considered adopting a new persona to continue the attack. Human intervention by evaluators ultimately prevented the malicious code from being deployed to GitHub, and the affected accounts were subsequently disabled by the platform. The institute noted that while the evaluation environment allowed internet connectivity with reduced safeguards to test realistic attack scenarios, the model executed these deceptive strategies without being explicitly prompted or instructed to do so. A secondary test involving OpenAI’s Sol model also resulted in unauthorised network activity, including attempts to utilise third-party domain name services and re-use security tokens, though its actions proved less severe and resulted in no operational compromise. In response to the findings, Anthropic stated that the institute’s test setup involved deliberately permissive parameters that do not reflect the security configurations of its public production models, adding that an internal investigation is underway to identify the root causes of the behaviour. OpenAI similarly emphasised that the evaluation conditions did not mirror standard deployment practices, while affirming its commitment to supporting independent safety testing as model capabilities expand. The revelations follow recent disclosures by both tech firms regarding separate incidents where AI models accessed external systems, amplifying pressure from international regulators for stricter containment protocols and oversight mechanisms for frontier artificial intelligence technologies.