NVIDIA has launched NVLink Fusion, integrating NVHBM (Custom High-Bandwidth Memory) to address the demands of trillion-parameter workloads in AI infrastructure. This technology aims to optimize the performance of AI systems...
Partner Spotlight
This space is available
Want to reach practitioners building with AI?
We'd love to hear from you.
Problem — The paper addresses the unexamined reliance on generative AI in cognitive tasks, leading to a gap between presented outputs and defensible knowledge. This phenomenon, termed ‘epistemic debt,’ emerges...
Problem The paper addresses the gap in existing AI frameworks that lack the ability to autonomously adapt to dynamic environments, drawing inspiration from biological organisms. It highlights the need for...
Problem This work addresses the gap in predicting cellular responses to untested drugs, particularly in the context of cancer therapeutics. Existing models often lack the ability to generalize effectively to...
OpenAI’s first in-house inference chip, Jalapeño, has reportedly outperformed Nvidia’s Blackwell and Rubin chips in both throughput and energy efficiency, as tested by SemiAnalysis. Dylan Patel, CEO of SemiAnalysis, noted...
{ “meta”: “This paper establishes the W[1]-hardness of computing $L_p$-Lipschitz constants for two-layer ICNNs, resolving an open problem in parameterized complexity.”, “body”: “## Problem\nThe paper addresses the challenge of computing...
Problem This preprint addresses the gap in understanding how AI professionals make sense of the rapid advancements in artificial intelligence and the societal implications that arise from these developments. The...
NVIDIA is showcasing its RTX Spark technology at the Gamescom conference this week in Cologne, Germany. Major game publishers including Electronic Arts, Ubisoft, and developer Embark are bringing blockbuster titles...
The Hugging Face Blog provides an overview of the Granite 4.2 large language models (LLMs) developed by IBM, highlighting their architectural innovations and performance enhancements. The article emphasizes that Granite...
Nvidia’s Groq 3 LPX inference chip claims a performance of 3,400 tokens per second on the Gemma 4 31B model, asserting it is four times faster than Cerebras. However, achieving...
The UK is the first country to gain access to Ukraine’s Avengers Labs, which features a dataset of 5 million annotated combat images. Three British startups are collaborating on pilot...
Chinese state-backed hacking groups have more than doubled their attacks since they began using AI models like DeepSeek to write exploit code and scan networks, according to TeamT5. This increase...
OpenAI’s CFO Sarah Friar stated that advancements in chips, compute, models, and products are compounding to deliver more useful intelligence at greater scale and lower cost. This reflects a strategic...
OpenAI’s Jalapeño chip delivers faster, more power-efficient AI inference, achieving higher throughput and lower latency for modern models. This advancement positions Jalapeño as a leader in the AI inference chip...
OpenAI has banned accounts originating from Russia that were part of a covert influence campaign. These accounts were reportedly using AI to promote a fake Israel-based think tank and a...
AI factories require continuous operation to generate intelligence at scale. NVIDIA emphasizes that XPUs, custom accelerators designed for AI-native companies, are crucial for optimizing performance metrics such as tokens per...
NVIDIA has announced the full production of Groq 3 LPX, enhancing the Vera Rubin NVL72 AI inference system. This development aims to improve fast token generation for agentic systems, crucial...
Intel introduced its architectures for Agentic AI at Hot Chips 2026, emphasizing a fundamental shift in computing design and delivery. Key components include the Diamond Rapids processor and the Crescent...
Problem — This preprint addresses the gap in understanding why Large Language Models (LLMs) sometimes fail to exhibit ethical behavior, particularly in scenarios where their objectives of helpfulness and harmlessness...
Thomson Reuters Invests $40M to Develop Proprietary AI Model ‘Thomson’
An investigation found that 17% of responses from AI chatbots, including ChatGPT, Gemini, Grok, and Claude, linked users to the anti-abortion organization Profemina. This raises concerns about the lack of...
Nvidia is negotiating an investment in Perplexity, which is currently valued at over $30 billion, representing a more than 50% increase from its last funding round. Perplexity has reported annualized...
OCR It is a Chrome extension that utilizes the Tesseract OCR engine for text extraction from un-copyable documents. Users can start or stop the automatic run with the hotkey ⌥⇧A,...
Problem — This work addresses the gap in efficient data compression methods for scientific datasets, particularly focusing on preserving fine details and signal fidelity. The authors propose a novel approach,...
Problem This work addresses the challenge of compressing large-scale scientific measurements into compact representations while preserving fine-scale details. The authors highlight a gap in existing methods that struggle to maintain...
Problem This work addresses the gap in existing exoskeleton control systems, which often rely on task-specific programming and lack adaptability to varying user needs. The authors highlight the need for...
