Agentic AI

195 articles in this category (Page 5 of 9)

AI NewsAgentic AIAI Paper Summary

Securing Autonomous LLM Agents: Tsinghua and Ant Group Unveil a Five-Layer Security Framework for OpenClaw

Tsinghua and Ant Group researchers reveal that 26% of community-contributed tools for LLM agents contain security vulnerabilities, requiring a new five-layer framework.

Mar 18, 2026

AI NewsAgentic AIAI Governance

Building Enterprise AI Governance with OpenClaw Gateway and Policy Engines

Implement a robust AI governance layer using OpenClaw to classify risks, enforce human-in-the-loop approvals for moderate-impact tasks, and maintain auditable execution traces for autonomous agents.

Mar 15, 2026

AI NewsAgentic AIDatabases

OpenViking: A Hierarchical Filesystem-Based Context Database for AI Agents

OpenViking introduces a virtual filesystem for AI memory, reducing token consumption from 24.6M to 4.2M while increasing task completion rates to 52.08% on the LoCoMo10 dataset.

Mar 15, 2026

AI NewsAgentic AIOpen Source

Garry Tan Releases gstack: An Open-Source Workflow Layer for Claude Code

Garry Tan releases gstack, an open-source toolkit providing 8 opinionated workflow skills for Claude Code, featuring a persistent Chromium daemon to reduce tool latency.

Mar 14, 2026

AI NewsAgentic AIArtificial Intelligence

Google DeepMind's Aletheia: Bridging Competitive Math and Autonomous Research

DeepMind's Aletheia agent achieves 95.1% accuracy on IMO-Proof Bench Advanced, automating PhD-level research and resolving 4 Erdős open questions.

Mar 13, 2026

AI NewsAgentic AIAI Agents

Model Context Protocol (MCP) vs. AI Agent Skills: A Deep Dive into Structured Tools and Behavioral Guidance for LLMs

A technical comparison of MCP's standardized tool interfaces and Skills' natural-language behavioral guidance for scaling AI agent capabilities and external system integration.

Mar 13, 2026

AI NewsAgentic AIOpen Source

Stanford's OpenJarvis: A Local-First Framework for On-Device Personal AI Agents

Stanford releases OpenJarvis, a local-first AI framework that handles 88.7% of reasoning queries on-device with 5.3x intelligence efficiency gains.

Mar 12, 2026

AI NewsAgentic AIAI Agents

Designing Streaming Decision Agents for Dynamic Environments

Build a Streaming Decision Agent using online A* and receding-horizon control to handle dynamic obstacles with a 0.85 risk gate for reactive adaptation.

Mar 11, 2026

AI NewsAgentic AILarge Language Model

NVIDIA Nemotron 3 Super: 120B Parameter Hybrid MoE Model for Agentic AI

NVIDIA's Nemotron 3 Super is a 120B parameter hybrid Mamba-Attention MoE model delivering 5x higher throughput for complex agentic AI applications.

Mar 11, 2026

AI NewsAgentic AIAI Infrastructure

NVIDIA Nemotron-Terminal: Scaling LLM Agents with Systematic Data Engineering

NVIDIA releases Nemotron-Terminal, a 32B model that outperforms the 480B Qwen3-Coder on terminal benchmarks using the Terminal-Task-Gen pipeline.

Mar 10, 2026

AI NewsVoice AIAgentic AI

Fish Audio S2-Pro: High-Fidelity TTS with Dual-AR Architecture and Sub-150ms Latency

Fish Audio S2-Pro introduces a Dual-AR framework and Residual Vector Quantization to deliver 44.1kHz speech synthesis with 100ms latency on NVIDIA H200.

Mar 10, 2026

AI NewsAgentic AIAI Agents

Building Self-Designing Meta-Agents for Automated AI Architecture Construction

Michal Sutter details a Meta-Agent framework that automatically constructs and refines task-specific AI agents using dynamic tool selection and self-improvement loops.

Mar 10, 2026

AI NewsAgentic AISoftware Engineering

Anthropic Claude Code: Automating Complex Security Research with Agentic Reasoning

Anthropic launches Claude Code featuring agentic loops capable of 21.2 tool calls per task, identifying 14 high-severity Firefox vulnerabilities in two weeks.

Mar 9, 2026

AI NewsAgentic AIOpen Source

ByteDance Releases DeerFlow 2.0: Open-Source SuperAgent Harness for Complex Tasks

ByteDance releases DeerFlow 2.0, an open-source SuperAgent framework that executes tasks in isolated Docker containers to build websites and automate data pipelines.

Mar 9, 2026

AI NewsAgentic AIAI Agents

Building Risk-Aware AI Agents with Internal Critics and Uncertainty Estimation

Develop reliable AI agents using internal critics and uncertainty estimation to quantify risk through entropy and consistency scores, ensuring robust decision-making in production environments.

Mar 9, 2026

AI NewsAgentic AIAI Infrastructure

Andrej Karpathy Open-Sources 'Autoresearch': A 630-Line Tool for Autonomous ML Experiments

Andrej Karpathy released autoresearch, a 630-line Python tool enabling AI agents to autonomously optimize ML models on single GPUs, achieving a 19% validation improvement in real-world tests.

Mar 8, 2026

AI NewsAgentic AIAI Agents

Building Next-Gen Agentic AI: A Framework for Cognitive Blueprint Runtime Agents

Build cognitive blueprint-driven AI agents that plan, execute, and validate tasks using a modular runtime engine and Pydantic-based structured memory.

Mar 7, 2026

AI NewsSoftware EngineeringAgentic AI

Google AI Releases Android Bench: Specialized Evaluation for Mobile LLMs

Google AI releases Android Bench, an open-source framework where Gemini 3.1 Pro Preview achieved a top 72.4% success rate on real-world Android tasks.

Mar 6, 2026

AI NewsAgentic AISoftware Engineering

OpenAI Introduces Codex Security: Context-Aware Vulnerability Detection and Patching

OpenAI launches Codex Security, an agentic tool that reduced security noise by 84% in beta testing across 1.2 million commits.

Mar 6, 2026

AI NewsAgentic AISoftware Engineering

Google AI Releases gws CLI for Unified Workspace API Management

Google AI has launched gws, an open-source CLI tool providing a unified interface for Workspace APIs like Drive and Gmail, featuring native Model Context Protocol (MCP) support for AI agents.

Mar 5, 2026

AI NewsAgentic AITutorials

Designing Advanced Tree-of-Thoughts Agents for Multi-Branch LLM Reasoning

Build a Tree-of-Thoughts reasoning agent using FLAN-T5 that solves complex 24-game puzzles through beam search and heuristic scoring.

Mar 5, 2026

AI NewsAgentic AIOpen Source

OpenAI Releases Symphony: An Open-Source Framework for Orchestrating Autonomous AI Coding Agents

OpenAI launches Symphony, an open-source Elixir-based framework for orchestrating autonomous AI agents through structured implementation runs and issue tracker polling.

Mar 5, 2026

AI NewsAgentic AILarge Language Model

Liquid AI Launches LocalCowork: Privacy-First Agent Workflows with LFM2-24B-A2B

Liquid AI releases LocalCowork and LFM2-24B-A2B, enabling local agentic workflows with 385ms tool-selection latency and a 14.5 GB memory footprint on consumer hardware.

Mar 5, 2026

AI NewsAgentic AIAI Agents

Build a Persistent AI Agent OS with Hierarchical Memory and FAISS Retrieval

Learn to build an EverMem-style AI OS using FAISS and SQLite for persistent memory, featuring automated consolidation and importance scoring to maintain context.

Mar 4, 2026