Full-text search across 26939 articles. Combine with topic and date filters; results sorted by relevance.
Alibaba's Qwen team released Qwen3.8-27B, a dense, natively multimodal 27B model under Apache 2.0 with weights on Hugging Face, designed for single-GPU agentic work. The model features a 3:1 hybrid of Gated DeltaNet line…
IBM Corp. announced a broad alliance with OpenAI to integrate OpenAI's GPT-5.6, Codex, and ChatGPT Work into IBM Consulting Advantage, targeting regulated sectors and launching a dedicated OpenAI Practice with tens of th…
Indian IT stocks have recovered most of their first-half losses, with the Nifty IT index halving its year-to-date decline from 32% on July 1 to roughly 17%, as investors treat the sector as a cheap hedge against overheat…
DeepSeek released DeepSeek Harness, an open-source agent framework under an MIT license, on August 13, 2026, the same day it declared V4-Pro generally available. The framework, which powers DeepSeek's own benchmark metho…
French startup Kog claims its software can deliver 30x faster LLM inference on standard datacenter GPUs, demoing 3,000 tokens per second on a 2-billion-parameter model using AMD MI300X and NVIDIA H200 chips. CEO Gaël Del…
SpaceX closed its $60 billion acquisition of AI coding startup Cursor on Friday, following an initial partnership in April and a commitment in June. The two companies have already collaborated on training Grok 4.5 and Gr…
Larsen & Toubro has rebranded its data center unit as Larsen & Toubro-Vyoma and partnered with Nvidia to build gigawatt-scale AI factory infrastructure in India, aiming to own part of the compute layer. Nvidia has named …
CrawlForge MCP Server v5.0.0 addresses a critical SSRF vulnerability that allowed IP-literal URLs to bypass hostname-based guards, potentially exposing loopback, link-local, and cloud metadata endpoints. The update also …
DeepSeek released a developer preview of Harness, a software framework for building autonomous AI agents, on Thursday, August 14, 2026. The framework offers four operational modes—standard, code-focused, creative, and mi…
Taiwan's Ministry of Digital Affairs reported that an AI-assisted cyberattack targeted government agencies in July, with affected bodies completing incident handling, while reports from The Register, The Guardian, and Te…
Agility Robotics Inc. announced on June 24 that it will go public through a merger with Churchill Capital Corp XI, a SPAC, at a $2.5 billion pre-money valuation. The deal is expected to generate about $620 million in gro…
A developer's comparison of vLLM and Ollama for LLM serving in 2026 shows that while Ollama is simpler for single-user local use, vLLM outperforms it dramatically under concurrency, with throughput up to nearly 20x highe…
Mistral OCR 4.1, released by Mistral AI, adds block-level confidence scores via a new `confidence_scores_granularity` parameter, enabling per-block routing to human review or indexing. The update, which aliases `mistral-…
HD Moore, founder and CEO of runZero, will present at Black Hat USA 2026 on August 5, revealing that 86,000 server backdoor chips are exposed, based on his research into baseboard management controllers (BMCs) and IPMI v…
AgentsKit, a new developer tool, launches with 89 specialist agents, 103 skills, and 181 slash commands in a single npx command, enabling one person to plan, build, test, ship, and market a product using Claude Code. The…
Bitcoin Red Team, a voluntary research collective, has adopted Chinese AI models, including Moonshot AI's 'Kimi K3,' for open-source security audits after U.S. tools imposed strict guardrails, and in August scanned 390 p…
A developer troubleshooting the GitHub MCP server in Claude Desktop found that all 26 tools failed with 'Tool execution failed' in under a second, despite being listed correctly. After initially suspecting an expired tok…
Baruch Sadogursky and Patrick Debois presented at a conference that stuffing a large CLAUDE.md file into a coding agent's context window caused an AI coding assistant to fail a test, returning a 500 error instead of the …
Vault Operator, an AI agent for Obsidian, operates inside a user's knowledge base to read notes, follow links, and act on behalf of the user, with features including inline AI chat, block-level provenance, web clipping, …
The EU AI Act's Article 50, enforceable on August 2, 2026, requires providers of generative AI systems to mark synthetic audio, image, video, and text outputs in a machine-readable format detectable as artificially gener…