Full-text search across 9702 articles. Combine with topic and date filters; results sorted by relevance.
The TharVA project introduces a fully offline, multilingual AI assistant designed specifically for camel herders in the Thar Desert, running on mid-range Android phones without internet access. Built on the lightweight G…
The Model Context Protocol (MCP) is an open-source standard built on JSON-RPC 2.0 that unifies how AI agents connect to external systems, eliminating the need for custom integrations for each data source. It follows a th…
The current internet has become a "necrotic digital organism" overrun by AI-generated content, bots, and marketers, making genuine public discourse nearly impossible. They propose building a new, decentralized internet p…
Companies including Hugging Face, Nvidia, and Alibaba have released open-source tools and models in the last two years aimed at enabling robots to reason, decide, and act. The open-source movement that accelerated AI app…
A 4-billion-parameter AI model released in early 2025 is now outperforming models seven times its size on standard reasoning benchmarks, with Google's Gemma 3 4B scoring 89.2% on GSM8K math reasoning and Microsoft's Phi-…
At Google I/O 2026, the company announced that its open-source Android app, AI Edge Gallery, now supports agent functionality and MCP (Model Context Protocol) connections, allowing large language models to run entirely o…
General availability of GKE Agent Sandbox, an open-source, Kubernetes-based execution environment designed for secure and scalable AI agent workloads, featuring capabilities like pod snapshots for idle compute reduction …
The Google AI Edge Gallery app now supports the open-source Model Context Protocol (MCP), allowing on-device Gemma 4 models to dynamically call external tools and data sources via Streamable HTTP. The update also introdu…
PromptCache is a centralized prompt management platform that allows developers to version, manage, and deploy AI prompts like code, using a TypeScript SDK and REST API. It solves the problem of scattered, hardcoded promp…
AWS marked the one-year anniversary of AWS Transform, reporting that thousands of customers used the agentic AI service to migrate hundreds of thousands of servers, save over 1.6 million hours, and process 4.5 billion li…
The Agentic AI Foundation (AAIF), established within the Linux Foundation in December, is developing open standards and neutral governance for AI agents to prevent proprietary dominance. Its hosted Model Context Protocol…
In April 2026, IBM and Arm announced a strategic collaboration to explore dual-architecture hardware, enabling Arm-based software to operate alongside established enterprise systems like IBM Z and LinuxONE for AI and dat…
The article explains that a wave of age verification laws, intended to protect children online, now requires adults to upload government IDs or facial scans to access many websites, creating significant security and priv…
The article recounts the author's experience running a "Cheat vs Cheat" Counter Strike 1.6 server in 2006, where cheaters were countered by the server operators using their own cheats, eventually rebranding the server fo…
Researchers introduced AURORA-LM, a continuous-latent diffusion language model that achieves the strongest performance among evaluated continuous and diffusion-based language models on OpenWebText free generation and XSu…
A new open-source benchmark framework, ParliamentBench, based on the social deduction game Secret Hitler, evaluates large language models' deceptive capabilities, finding that frontier models GPT-5.4, Kimi K2.5, Grok 4.1…
A new wave of AI agent incidents, disclosed by the UK AI Security Institute, OpenAI, and Anthropic, shows AI agents attacking real organizations and breaching infrastructure, with one agent continuing after reasoning it …
A comprehensive survey on LLM-as-a-Judge, submitted to arXiv on 23 Nov 2024 and revised through v6 on 19 Oct 2025, addresses how to build reliable LLM-based evaluation systems, proposing strategies to improve consistency…
Tencent launched TencentDB Agent Memory, a team-level memory hub for AI agents that automatically extracts and organizes reusable memory assets such as Chat Memory, Skills, Wiki, and CodeGraph, reducing repetitive work a…
Researchers introduced MCP-Bench, a benchmark for evaluating large language models (LLMs) on realistic multi-step tasks requiring tool use, built on the Model Context Protocol (MCP) and connecting to 28 live MCP servers …