OpenAI, the AI research company behind ChatGPT, has secured a staggering $40 billion in funding led by SoftBank, valuing the company at an unprecedented $300 billion, enabling it to accelerate AI research, expand compute infrastructure, and bolster its Stargate joint venture with Microsoft.
Transformer models, neural network architectures employing self-attention mechanisms, have revolutionized natural language processing tasks, achieving remarkable performance in machine translation, question answering, and text generation through models like BERT, GPT, and ChatGPT.
In a groundbreaking move, Nvidia has teamed up with Cassava Technologies to establish Africa's pioneering AI factory, offering 'AI as a service' across the continent by harnessing fiber-optic networks and energy-efficient data centers, addressing the region's computing power shortage and fostering its participation in the AI revolution.
Researchers developed an end-to-end neural network trained with reinforcement learning, enabling robots to learn human-like walking in simulation and seamlessly transfer this capability to real hardware without additional tuning, achieving robust and natural locomotion across multiple robots.
Extropic's pioneering thermodynamic chips leverage energy fluctuations for ultra-efficient probabilistic computing, promising a 3-4 order magnitude boost in efficiency over current hardware and a potential revolution in energy-efficient AI and simulations.
OpenAI's groundbreaking o3 model, combining deep learning and program search, shattered records with a remarkable 75.7% score on the ARC-AGI benchmark, showcasing unprecedented task adaptation capabilities, yet falling short of true AGI as tougher challenges loom.
Google unveils Gemini 2.5, a groundbreaking AI model that 'thinks' before answering, with Gemini 2.5 Pro Experimental being its most intelligent multimodal reasoning AI yet, aiming to outpace rivals OpenAI and Anthropic in the AI race.
Intensifying the global AI race, Chinese startup DeepSeek unveiled a powerful upgrade to its language model, DeepSeek-V3-0324, rivaling OpenAI and Anthropic with enhanced reasoning and coding capabilities while offering cost advantages.
In a surprising move, Microsoft declined a nearly $12 billion deal for data center capacity, allowing OpenAI to swiftly secure the contract, highlighting the intense competition and high stakes in the AI industry as CoreWeave's impending IPO is viewed as a bellwether for the sector.
Nvidia unveiled formidable new AI chips and personal 'supercomputers' at GTC 2025, projecting confidence amid headwinds like tariffs and competition, while exploring quantum computing and positioning its hardware for AI inference demand despite concerns about efficient models.