How Nvidia changed the open source AI game with Nemotron 3

Rivian has announced the replacement of Nvidia with its own AI chip, named RAP1, as part of its strategy to achieve Level 4 autonomy in its vehicles. This decision coincides with a partnership with Volkswagen, indicating a significant shift in Rivian's approach to autonomous driving technology.

Nvidia is reportedly considering a production cut of 30 to 40% for its GeForce RTX 50 series GPUs in the first half of 2026, primarily due to ongoing memory shortages. This decision, based on unconfirmed reports from Board Channels, suggests that availability of these graphics cards may become as challenging as it was at their initial launch.

Nvidia has introduced the Nemotron 3 family, which integrates Mamba and Transformer architectures to efficiently manage long context windows for AI agents. This hybrid approach aims to optimize resource usage while enhancing performance in AI applications.

The recent interactions between Silicon Valley executives and British politicians highlight a growing trend where former political leaders, such as George Osborne, Nick Clegg, and Tony Blair, are taking on roles in tech firms. This shift was exemplified by Nvidia CEO Jensen Huang's event in London during Donald Trump's state visit, where he promoted AI's potential and announced significant investments in the UK.

Nvidia has acquired SchedMD, the leading developer of the Slurm workload manager, as part of its strategy to enhance its open-source AI portfolio. This acquisition aligns with Nvidia's ongoing efforts to strengthen its high-performance computing (HPC) capabilities in the rapidly evolving AI landscape.