A developer has implemented a UEFI application that runs LLM inference directly from boot without any operating system or kernel, using zero-dependency C code for the entire stack from tokenizer to inference engine.
Google has launched Nano Banana 2, a new AI image generation model combining Pro-level capabilities with Flash-level speed, featuring advanced world knowledge and subject consistency.
Anthropic CEO Dario Amodei confirmed in a CBS interview that the company built custom Claude models for the military that are 1-2 generations ahead of consumer versions, deployed on classified cloud infrastructure.
NVIDIA revealed detailed specs for Vera Rubin NVL72. Each Rubin GPU delivers 50 PFLOPS inference (5x Blackwell GB200), 22 TB/s HBM4 bandwidth (2.8x Blackwell), and cuts inference cost per million tokens by 10x. Ships H2 2026.
The India AI Impact Summit drew pledges exceeding $250 billion, with Adani Group committing $100B for renewable-powered AI data centers and Reliance Industries pledging $110B over seven years. Microsoft also committed $50B for Global South AI infrastructure.
OpenAI closed a $110B round led by Amazon ($50B), Nvidia ($30B), and SoftBank ($30B). ChatGPT now reaches 900 million weekly active users and 50 million paying subscribers.
AI researcher Andrej Karpathy argues that programming has fundamentally changed over the last two months, particularly since December when coding agents started actually working. Developers are shifting from writing code to directing and managing AI agents in parallel.
OpenAI CEO Sam Altman announced a Pentagon deal to deploy AI models in classified networks just hours after Anthropic was blacklisted by the Trump administration. The agreement explicitly includes prohibitions on mass domestic surveillance and autonomous weapons.
OpenAI CEO Sam Altman announced a Pentagon deal to deploy AI models in classified networks just hours after Anthropic was blacklisted by the Trump administration. The agreement explicitly includes prohibitions on mass domestic surveillance and autonomous weapons.
NVIDIA announced new AI Blueprint workflows for telecom on February 28, 2026, combining Nemotron reasoning models with NVIDIA NIM microservices. The company says early partners including Amdocs, BubbleRAN, and ServiceNow are applying the stack to network configuration and optimization.
Google DeepMind announced SIMA 2 on November 13, 2025 as a generalist foundation model for virtual 3D environments. The system is designed to play and reason alongside humans, with in-context learning that can improve behavior from examples.
Anthropic announced Responsible Scaling Policy v3 on February 24, 2026 and paired it with a Frontier Safety Roadmap. The company says it will update the policy every 3-6 months and publish model-specific Risk Reports to improve verifiability.