Neocloud Lambda, an infrastructure firm, has raised $1bn in private debt to purchase Nvidia AI chips for leasing to Microsoft. The financing underscores the capital intensity of the artificial-intelligence boom, as companies race to secure cutting-edge hardware to meet surging demand from cloud customers. Source
A federal judge ruled that the Trump administration illegally designated Anthropic, an AI safety company, as a supply-chain risk, handing the startup a court victory. The ruling comes as Anthropic pursues a second lawsuit challenging Pentagon restrictions on its business with the Department of Defence in Washington. Source
Sandhya Devanathan, a Meta executive, is joining OpenAI to oversee operations across Southeast Asia and Australia. The departure comes as Meta faces intensifying scrutiny of its business practices in India and other markets in the region. Source
Researchers at Anthropic have demonstrated self-improving AI systems that were able to improve performance on all 10 benchmarks measuring specific misaligned behaviors without degrading overall performance. The work offers a concrete example of automated systems autonomously addressing their own failure modes. Source
OpenAI has decided to wind down its contract providing models to Cursor, a code-completion tool, following Cursor's acquisition by SpaceX. The move reflects potential conflicts between OpenAI and Elon Musk's companies over artificial-intelligence technology and market positioning. Source